{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Defaulting to user installation because normal site-packages is not writeable\n",
      "Collecting torchtext\n",
      "  Downloading torchtext-0.15.2-cp310-cp310-manylinux1_x86_64.whl (2.0 MB)\n",
      "\u001b[2K     \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.0/2.0 MB\u001b[0m \u001b[31m1.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m[36m0:00:01\u001b[0m[36m0:00:01\u001b[0m:01\u001b[0m\n",
      "\u001b[?25hCollecting tqdm\n",
      "  Using cached tqdm-4.65.0-py3-none-any.whl (77 kB)\n",
      "Requirement already satisfied: numpy in /home/gedin/.local/lib/python3.10/site-packages (from torchtext) (1.24.3)\n",
      "Collecting torchdata==0.6.1\n",
      "  Downloading torchdata-0.6.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (4.6 MB)\n",
      "\u001b[2K     \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m4.6/4.6 MB\u001b[0m \u001b[31m1.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0mm eta \u001b[36m0:00:01\u001b[0m[36m0:00:01\u001b[0m\n",
      "\u001b[?25hRequirement already satisfied: requests in /usr/lib/python3/dist-packages (from torchtext) (2.25.1)\n",
      "Collecting torch==2.0.1\n",
      "  Downloading torch-2.0.1-cp310-cp310-manylinux1_x86_64.whl (619.9 MB)\n",
      "\u001b[2K     \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m619.9/619.9 MB\u001b[0m \u001b[31m1.1 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0mm eta \u001b[36m0:00:01\u001b[0m[36m0:00:09\u001b[0m\n",
      "\u001b[?25hCollecting sympy\n",
      "  Downloading sympy-1.12-py3-none-any.whl (5.7 MB)\n",
      "\u001b[2K     \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m5.7/5.7 MB\u001b[0m \u001b[31m1.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0mm eta \u001b[36m0:00:01\u001b[0m[36m0:00:01\u001b[0m\n",
      "\u001b[?25hCollecting nvidia-cudnn-cu11==8.5.0.96\n",
      "  Using cached nvidia_cudnn_cu11-8.5.0.96-2-py3-none-manylinux1_x86_64.whl (557.1 MB)\n",
      "Collecting nvidia-cuda-cupti-cu11==11.7.101\n",
      "  Using cached nvidia_cuda_cupti_cu11-11.7.101-py3-none-manylinux1_x86_64.whl (11.8 MB)\n",
      "Collecting nvidia-cusparse-cu11==11.7.4.91\n",
      "  Using cached nvidia_cusparse_cu11-11.7.4.91-py3-none-manylinux1_x86_64.whl (173.2 MB)\n",
      "Collecting networkx\n",
      "  Using cached networkx-3.1-py3-none-any.whl (2.1 MB)\n",
      "Collecting nvidia-cufft-cu11==10.9.0.58\n",
      "  Using cached nvidia_cufft_cu11-10.9.0.58-py3-none-manylinux1_x86_64.whl (168.4 MB)\n",
      "Collecting filelock\n",
      "  Downloading filelock-3.12.0-py3-none-any.whl (10 kB)\n",
      "Collecting nvidia-cuda-runtime-cu11==11.7.99\n",
      "  Using cached nvidia_cuda_runtime_cu11-11.7.99-py3-none-manylinux1_x86_64.whl (849 kB)\n",
      "Collecting triton==2.0.0\n",
      "  Downloading triton-2.0.0-1-cp310-cp310-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (63.3 MB)\n",
      "\u001b[2K     \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m63.3/63.3 MB\u001b[0m \u001b[31m1.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0mm eta \u001b[36m0:00:01\u001b[0m[36m0:00:02\u001b[0m\n",
      "\u001b[?25hCollecting nvidia-cusolver-cu11==11.4.0.1\n",
      "  Using cached nvidia_cusolver_cu11-11.4.0.1-2-py3-none-manylinux1_x86_64.whl (102.6 MB)\n",
      "Requirement already satisfied: jinja2 in /home/gedin/.local/lib/python3.10/site-packages (from torch==2.0.1->torchtext) (3.1.2)\n",
      "Collecting nvidia-cublas-cu11==11.10.3.66\n",
      "  Using cached nvidia_cublas_cu11-11.10.3.66-py3-none-manylinux1_x86_64.whl (317.1 MB)\n",
      "Collecting typing-extensions\n",
      "  Downloading typing_extensions-4.6.3-py3-none-any.whl (31 kB)\n",
      "Collecting nvidia-nccl-cu11==2.14.3\n",
      "  Using cached nvidia_nccl_cu11-2.14.3-py3-none-manylinux1_x86_64.whl (177.1 MB)\n",
      "Collecting nvidia-cuda-nvrtc-cu11==11.7.99\n",
      "  Using cached nvidia_cuda_nvrtc_cu11-11.7.99-2-py3-none-manylinux1_x86_64.whl (21.0 MB)\n",
      "Collecting nvidia-curand-cu11==10.2.10.91\n",
      "  Using cached nvidia_curand_cu11-10.2.10.91-py3-none-manylinux1_x86_64.whl (54.6 MB)\n",
      "Collecting nvidia-nvtx-cu11==11.7.91\n",
      "  Using cached nvidia_nvtx_cu11-11.7.91-py3-none-manylinux1_x86_64.whl (98 kB)\n",
      "Requirement already satisfied: urllib3>=1.25 in /usr/lib/python3/dist-packages (from torchdata==0.6.1->torchtext) (1.26.5)\n",
      "Requirement already satisfied: wheel in /usr/lib/python3/dist-packages (from nvidia-cublas-cu11==11.10.3.66->torch==2.0.1->torchtext) (0.37.1)\n",
      "Requirement already satisfied: setuptools in /usr/lib/python3/dist-packages (from nvidia-cublas-cu11==11.10.3.66->torch==2.0.1->torchtext) (59.6.0)\n",
      "Collecting lit\n",
      "  Downloading lit-16.0.5.tar.gz (138 kB)\n",
      "\u001b[2K     \u001b[38;2;114;156;31m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m138.0/138.0 KB\u001b[0m \u001b[31m1.7 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m[31m1.6 MB/s\u001b[0m eta \u001b[36m0:00:01\u001b[0m\n",
      "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25ldone\n",
      "\u001b[?25hCollecting cmake\n",
      "  Using cached cmake-3.26.3-py2.py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (24.0 MB)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /usr/lib/python3/dist-packages (from jinja2->torch==2.0.1->torchtext) (2.0.1)\n",
      "Collecting mpmath>=0.19\n",
      "  Using cached mpmath-1.3.0-py3-none-any.whl (536 kB)\n",
      "Building wheels for collected packages: lit\n",
      "  Building wheel for lit (setup.py) ... \u001b[?25ldone\n",
      "\u001b[?25h  Created wheel for lit: filename=lit-16.0.5-py3-none-any.whl size=88192 sha256=f6c57a31a147cbfe0af3d6bf4b856390ad14c28a9ddb38c8044ec29331b35c26\n",
      "  Stored in directory: /home/gedin/.cache/pip/wheels/eb/02/84/d82f0b1a6098209edf7e3607be6cc592ebbc015a8a3127c68d\n",
      "Successfully built lit\n",
      "Installing collected packages: mpmath, lit, cmake, typing-extensions, tqdm, sympy, nvidia-nvtx-cu11, nvidia-nccl-cu11, nvidia-cusparse-cu11, nvidia-curand-cu11, nvidia-cufft-cu11, nvidia-cuda-runtime-cu11, nvidia-cuda-nvrtc-cu11, nvidia-cuda-cupti-cu11, nvidia-cublas-cu11, networkx, filelock, nvidia-cusolver-cu11, nvidia-cudnn-cu11, triton, torch, torchdata, torchtext\n",
      "Successfully installed cmake-3.26.3 filelock-3.12.0 lit-16.0.5 mpmath-1.3.0 networkx-3.1 nvidia-cublas-cu11-11.10.3.66 nvidia-cuda-cupti-cu11-11.7.101 nvidia-cuda-nvrtc-cu11-11.7.99 nvidia-cuda-runtime-cu11-11.7.99 nvidia-cudnn-cu11-8.5.0.96 nvidia-cufft-cu11-10.9.0.58 nvidia-curand-cu11-10.2.10.91 nvidia-cusolver-cu11-11.4.0.1 nvidia-cusparse-cu11-11.7.4.91 nvidia-nccl-cu11-2.14.3 nvidia-nvtx-cu11-11.7.91 sympy-1.12 torch-2.0.1 torchdata-0.6.1 torchtext-0.15.2 tqdm-4.65.0 triton-2.0.0 typing-extensions-4.6.3\n"
     ]
    }
   ],
   "source": [
    "!pip install torchtext"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_file ='train/in.tsv.xz'\n",
    "test_file = 'test-A/in.tsv.xz'\n",
    "out_file = 'test-A/out.tsv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from itertools import islice\n",
    "import regex as re\n",
    "import sys\n",
    "from torchtext.vocab import build_vocab_from_iterator\n",
    "import lzma\n",
    "import pickle\n",
    "import re\n",
    "import torch\n",
    "from torch import nn\n",
    "from torch.utils.data import IterableDataset\n",
    "import itertools\n",
    "from torch.utils.data import DataLoader\n",
    "import gc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "embed_size = 300\n",
    "device = 'cuda'\n",
    "vocab_size = 25000\n",
    "batch_s = 3200\n",
    "learning_rate = 0.0001\n",
    "epochs = 4\n",
    "k = 20 #top k words\n",
    "wildcard_minweight = 0.001"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "###preprocessing\n",
    "def preprocess(line):\n",
    "    line = get_rid_of_header(line)\n",
    "    line = replace_endline(line)\n",
    "    return line\n",
    "\n",
    "def get_rid_of_header(line):\n",
    "    line = line.split('\\t')[6:]\n",
    "    return \"\".join(line)\n",
    "    \n",
    "def replace_endline(line):\n",
    "    line = line.replace(\"\\\\n\", \" \")\n",
    "    return line\n",
    "\n",
    "\n",
    "def get_last_word(text):\n",
    "    \"\"\"Return the last word of a string.\"\"\"\n",
    "    last_word = \"\"\n",
    "    for i in range(len(text)-1, -1, -1):\n",
    "        if text[i] == ' ':\n",
    "            return last_word[::-1].rstrip()\n",
    "        else:\n",
    "            last_word += text[i]\n",
    "    return last_word[::-1].rstrip()\n",
    "\n",
    "def get_first_word(text):\n",
    "    \"\"\"Return the first word of a string.\"\"\"\n",
    "    word = \"\"\n",
    "    for i in range(len(text)-1):\n",
    "        if text[i] == ' ':\n",
    "            return word\n",
    "        else:\n",
    "            word += text[i]\n",
    "    return word\n",
    "\n",
    "\n",
    "def get_words_from_line(line):\n",
    "    line = line.rstrip()\n",
    "    yield '<s>'\n",
    "    line = preprocess(line)\n",
    "    for t in line.split(' '):\n",
    "        yield t\n",
    "    yield '</s>'\n",
    "\n",
    "\n",
    "def get_word_lines_from_file(file_name):\n",
    "    n = 0\n",
    "    with lzma.open(file_name, 'r') as fh:\n",
    "        for line in fh:\n",
    "            n+=1\n",
    "            if n%1000==0:\n",
    "                print(n)\n",
    "            yield get_words_from_line(line.decode('utf-8'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000\n",
      "2000\n",
      "3000\n",
      "4000\n",
      "5000\n",
      "6000\n",
      "7000\n",
      "8000\n",
      "9000\n",
      "10000\n",
      "11000\n",
      "12000\n",
      "13000\n",
      "14000\n",
      "15000\n",
      "16000\n",
      "17000\n",
      "18000\n",
      "19000\n",
      "20000\n",
      "21000\n",
      "22000\n",
      "23000\n",
      "24000\n",
      "25000\n",
      "26000\n",
      "27000\n",
      "28000\n",
      "29000\n",
      "30000\n",
      "31000\n",
      "32000\n",
      "33000\n",
      "34000\n",
      "35000\n",
      "36000\n",
      "37000\n",
      "38000\n",
      "39000\n",
      "40000\n",
      "41000\n",
      "42000\n",
      "43000\n",
      "44000\n",
      "45000\n",
      "46000\n",
      "47000\n",
      "48000\n",
      "49000\n",
      "50000\n",
      "51000\n",
      "52000\n",
      "53000\n",
      "54000\n",
      "55000\n",
      "56000\n",
      "57000\n",
      "58000\n",
      "59000\n",
      "60000\n",
      "61000\n",
      "62000\n",
      "63000\n",
      "64000\n",
      "65000\n",
      "66000\n",
      "67000\n",
      "68000\n",
      "69000\n",
      "70000\n",
      "71000\n",
      "72000\n",
      "73000\n",
      "74000\n",
      "75000\n",
      "76000\n",
      "77000\n",
      "78000\n",
      "79000\n",
      "80000\n",
      "81000\n",
      "82000\n",
      "83000\n",
      "84000\n",
      "85000\n",
      "86000\n",
      "87000\n",
      "88000\n",
      "89000\n",
      "90000\n",
      "91000\n",
      "92000\n",
      "93000\n",
      "94000\n",
      "95000\n",
      "96000\n",
      "97000\n",
      "98000\n",
      "99000\n",
      "100000\n",
      "101000\n",
      "102000\n",
      "103000\n",
      "104000\n",
      "105000\n",
      "106000\n",
      "107000\n",
      "108000\n",
      "109000\n",
      "110000\n",
      "111000\n",
      "112000\n",
      "113000\n",
      "114000\n",
      "115000\n",
      "116000\n",
      "117000\n",
      "118000\n",
      "119000\n",
      "120000\n",
      "121000\n",
      "122000\n",
      "123000\n",
      "124000\n",
      "125000\n",
      "126000\n",
      "127000\n",
      "128000\n",
      "129000\n",
      "130000\n",
      "131000\n",
      "132000\n",
      "133000\n",
      "134000\n",
      "135000\n",
      "136000\n",
      "137000\n",
      "138000\n",
      "139000\n",
      "140000\n",
      "141000\n",
      "142000\n",
      "143000\n",
      "144000\n",
      "145000\n",
      "146000\n",
      "147000\n",
      "148000\n",
      "149000\n",
      "150000\n",
      "151000\n",
      "152000\n",
      "153000\n",
      "154000\n",
      "155000\n",
      "156000\n",
      "157000\n",
      "158000\n",
      "159000\n",
      "160000\n",
      "161000\n",
      "162000\n",
      "163000\n",
      "164000\n",
      "165000\n",
      "166000\n",
      "167000\n",
      "168000\n",
      "169000\n",
      "170000\n",
      "171000\n",
      "172000\n",
      "173000\n",
      "174000\n",
      "175000\n",
      "176000\n",
      "177000\n",
      "178000\n",
      "179000\n",
      "180000\n",
      "181000\n",
      "182000\n",
      "183000\n",
      "184000\n",
      "185000\n",
      "186000\n",
      "187000\n",
      "188000\n",
      "189000\n",
      "190000\n",
      "191000\n",
      "192000\n",
      "193000\n",
      "194000\n",
      "195000\n",
      "196000\n",
      "197000\n",
      "198000\n",
      "199000\n",
      "200000\n",
      "201000\n",
      "202000\n",
      "203000\n",
      "204000\n",
      "205000\n",
      "206000\n",
      "207000\n",
      "208000\n",
      "209000\n",
      "210000\n",
      "211000\n",
      "212000\n",
      "213000\n",
      "214000\n",
      "215000\n",
      "216000\n",
      "217000\n",
      "218000\n",
      "219000\n",
      "220000\n",
      "221000\n",
      "222000\n",
      "223000\n",
      "224000\n",
      "225000\n",
      "226000\n",
      "227000\n",
      "228000\n",
      "229000\n",
      "230000\n",
      "231000\n",
      "232000\n",
      "233000\n",
      "234000\n",
      "235000\n",
      "236000\n",
      "237000\n",
      "238000\n",
      "239000\n",
      "240000\n",
      "241000\n",
      "242000\n",
      "243000\n",
      "244000\n",
      "245000\n",
      "246000\n",
      "247000\n",
      "248000\n",
      "249000\n",
      "250000\n",
      "251000\n",
      "252000\n",
      "253000\n",
      "254000\n",
      "255000\n",
      "256000\n",
      "257000\n",
      "258000\n",
      "259000\n",
      "260000\n",
      "261000\n",
      "262000\n",
      "263000\n",
      "264000\n",
      "265000\n",
      "266000\n",
      "267000\n",
      "268000\n",
      "269000\n",
      "270000\n",
      "271000\n",
      "272000\n",
      "273000\n",
      "274000\n",
      "275000\n",
      "276000\n",
      "277000\n",
      "278000\n",
      "279000\n",
      "280000\n",
      "281000\n",
      "282000\n",
      "283000\n",
      "284000\n",
      "285000\n",
      "286000\n",
      "287000\n",
      "288000\n",
      "289000\n",
      "290000\n",
      "291000\n",
      "292000\n",
      "293000\n",
      "294000\n",
      "295000\n",
      "296000\n",
      "297000\n",
      "298000\n",
      "299000\n",
      "300000\n",
      "301000\n",
      "302000\n",
      "303000\n",
      "304000\n",
      "305000\n",
      "306000\n",
      "307000\n",
      "308000\n",
      "309000\n",
      "310000\n",
      "311000\n",
      "312000\n",
      "313000\n",
      "314000\n",
      "315000\n",
      "316000\n",
      "317000\n",
      "318000\n",
      "319000\n",
      "320000\n",
      "321000\n",
      "322000\n",
      "323000\n",
      "324000\n",
      "325000\n",
      "326000\n",
      "327000\n",
      "328000\n",
      "329000\n",
      "330000\n",
      "331000\n",
      "332000\n",
      "333000\n",
      "334000\n",
      "335000\n",
      "336000\n",
      "337000\n",
      "338000\n",
      "339000\n",
      "340000\n",
      "341000\n",
      "342000\n",
      "343000\n",
      "344000\n",
      "345000\n",
      "346000\n",
      "347000\n",
      "348000\n",
      "349000\n",
      "350000\n",
      "351000\n",
      "352000\n",
      "353000\n",
      "354000\n",
      "355000\n",
      "356000\n",
      "357000\n",
      "358000\n",
      "359000\n",
      "360000\n",
      "361000\n",
      "362000\n",
      "363000\n",
      "364000\n",
      "365000\n",
      "366000\n",
      "367000\n",
      "368000\n",
      "369000\n",
      "370000\n",
      "371000\n",
      "372000\n",
      "373000\n",
      "374000\n",
      "375000\n",
      "376000\n",
      "377000\n",
      "378000\n",
      "379000\n",
      "380000\n",
      "381000\n",
      "382000\n",
      "383000\n",
      "384000\n",
      "385000\n",
      "386000\n",
      "387000\n",
      "388000\n",
      "389000\n",
      "390000\n",
      "391000\n",
      "392000\n",
      "393000\n",
      "394000\n",
      "395000\n",
      "396000\n",
      "397000\n",
      "398000\n",
      "399000\n",
      "400000\n",
      "401000\n",
      "402000\n",
      "403000\n",
      "404000\n",
      "405000\n",
      "406000\n",
      "407000\n",
      "408000\n",
      "409000\n",
      "410000\n",
      "411000\n",
      "412000\n",
      "413000\n",
      "414000\n",
      "415000\n",
      "416000\n",
      "417000\n",
      "418000\n",
      "419000\n",
      "420000\n",
      "421000\n",
      "422000\n",
      "423000\n",
      "424000\n",
      "425000\n",
      "426000\n",
      "427000\n",
      "428000\n",
      "429000\n",
      "430000\n",
      "431000\n",
      "432000\n"
     ]
    }
   ],
   "source": [
    "vocab = build_vocab_from_iterator(\n",
    "    get_word_lines_from_file(train_file),\n",
    "    max_tokens = vocab_size,\n",
    "    specials = ['<unk>'])\n",
    "\n",
    "with open('filename.pickle', 'wb') as handle:\n",
    "    pickle.dump(vocab, handle, protocol=pickle.HIGHEST_PROTOCOL)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'vocab' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m/tmp/ipykernel_20466/3224446201.py\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mvocab\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mlookup_tokens\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;36m0\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m10\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m2000\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'vocab' is not defined"
     ]
    }
   ],
   "source": [
    "vocab.lookup_tokens([0, 1, 2, 10, 2000])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Definicja sieci\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Naszą prostą sieć neuronową zaimplementujemy używając frameworku PyTorch.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "class SimpleBigramNeuralLanguageModel(nn.Module):\n",
    "  def __init__(self, vocabulary_size, embedding_size):\n",
    "      super(SimpleBigramNeuralLanguageModel, self).__init__()\n",
    "      self.model = nn.Sequential(\n",
    "          nn.Embedding(vocabulary_size, embedding_size),\n",
    "          nn.Linear(embedding_size, vocabulary_size),\n",
    "          nn.Softmax()\n",
    "      )\n",
    "    \n",
    "  def forward(self, x):\n",
    "      return self.model(x)\n",
    "\n",
    "with open('filename.pickle','rb') as handle:\n",
    "    vocab = pickle.load(handle)\n",
    "\n",
    "vocab.set_default_index(vocab['<unk>'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Help on Vocab in module torchtext.vocab.vocab object:\n",
      "\n",
      "class Vocab(torch.nn.modules.module.Module)\n",
      " |  Vocab(vocab) -> None\n",
      " |  \n",
      " |  Base class for all neural network modules.\n",
      " |  \n",
      " |  Your models should also subclass this class.\n",
      " |  \n",
      " |  Modules can also contain other Modules, allowing to nest them in\n",
      " |  a tree structure. You can assign the submodules as regular attributes::\n",
      " |  \n",
      " |      import torch.nn as nn\n",
      " |      import torch.nn.functional as F\n",
      " |  \n",
      " |      class Model(nn.Module):\n",
      " |          def __init__(self):\n",
      " |              super().__init__()\n",
      " |              self.conv1 = nn.Conv2d(1, 20, 5)\n",
      " |              self.conv2 = nn.Conv2d(20, 20, 5)\n",
      " |  \n",
      " |          def forward(self, x):\n",
      " |              x = F.relu(self.conv1(x))\n",
      " |              return F.relu(self.conv2(x))\n",
      " |  \n",
      " |  Submodules assigned in this way will be registered, and will have their\n",
      " |  parameters converted too when you call :meth:`to`, etc.\n",
      " |  \n",
      " |  .. note::\n",
      " |      As per the example above, an ``__init__()`` call to the parent class\n",
      " |      must be made before assignment on the child.\n",
      " |  \n",
      " |  :ivar training: Boolean represents whether this module is in training or\n",
      " |                  evaluation mode.\n",
      " |  :vartype training: bool\n",
      " |  \n",
      " |  Method resolution order:\n",
      " |      Vocab\n",
      " |      torch.nn.modules.module.Module\n",
      " |      builtins.object\n",
      " |  \n",
      " |  Methods defined here:\n",
      " |  \n",
      " |  __contains__(self, token: str) -> bool\n",
      " |      Args:\n",
      " |          token: The token for which to check the membership.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Whether the token is member of vocab or not.\n",
      " |  \n",
      " |  __getitem__(self, token: str) -> int\n",
      " |      Args:\n",
      " |          token: The token used to lookup the corresponding index.\n",
      " |      \n",
      " |      Returns:\n",
      " |          The index corresponding to the associated token.\n",
      " |  \n",
      " |  __init__(self, vocab) -> None\n",
      " |      Initializes internal Module state, shared by both nn.Module and ScriptModule.\n",
      " |  \n",
      " |  __len__(self) -> int\n",
      " |      Returns:\n",
      " |          The length of the vocab.\n",
      " |  \n",
      " |  __prepare_scriptable__(self)\n",
      " |      Return a JITable Vocab.\n",
      " |  \n",
      " |  append_token(self, token: str) -> None\n",
      " |      Args:\n",
      " |          token: The token used to lookup the corresponding index.\n",
      " |      \n",
      " |      Raises:\n",
      " |          RuntimeError: If `token` already exists in the vocab\n",
      " |  \n",
      " |  forward(self, tokens: List[str]) -> List[int]\n",
      " |      Calls the `lookup_indices` method\n",
      " |      \n",
      " |      Args:\n",
      " |          tokens: a list of tokens used to lookup their corresponding `indices`.\n",
      " |      \n",
      " |      Returns:\n",
      " |          The indices associated with a list of `tokens`.\n",
      " |  \n",
      " |  get_default_index(self) -> Union[int, NoneType]\n",
      " |      Returns:\n",
      " |          Value of default index if it is set.\n",
      " |  \n",
      " |  get_itos(self) -> List[str]\n",
      " |      Returns:\n",
      " |          List mapping indices to tokens.\n",
      " |  \n",
      " |  get_stoi(self) -> Dict[str, int]\n",
      " |      Returns:\n",
      " |          Dictionary mapping tokens to indices.\n",
      " |  \n",
      " |  insert_token(self, token: str, index: int) -> None\n",
      " |      Args:\n",
      " |          token: The token used to lookup the corresponding index.\n",
      " |          index: The index corresponding to the associated token.\n",
      " |      Raises:\n",
      " |          RuntimeError: If `index` is not in range [0, Vocab.size()] or if `token` already exists in the vocab.\n",
      " |  \n",
      " |  lookup_indices(self, tokens: List[str]) -> List[int]\n",
      " |      Args:\n",
      " |          tokens: the tokens used to lookup their corresponding `indices`.\n",
      " |      \n",
      " |      Returns:\n",
      " |          The 'indices` associated with `tokens`.\n",
      " |  \n",
      " |  lookup_token(self, index: int) -> str\n",
      " |      Args:\n",
      " |          index: The index corresponding to the associated token.\n",
      " |      \n",
      " |      Returns:\n",
      " |          token: The token used to lookup the corresponding index.\n",
      " |      \n",
      " |      Raises:\n",
      " |          RuntimeError: If `index` not in range [0, itos.size()).\n",
      " |  \n",
      " |  lookup_tokens(self, indices: List[int]) -> List[str]\n",
      " |      Args:\n",
      " |          indices: The `indices` used to lookup their corresponding`tokens`.\n",
      " |      \n",
      " |      Returns:\n",
      " |          The `tokens` associated with `indices`.\n",
      " |      \n",
      " |      Raises:\n",
      " |          RuntimeError: If an index within `indices` is not int range [0, itos.size()).\n",
      " |  \n",
      " |  set_default_index(self, index: Union[int, NoneType]) -> None\n",
      " |      Args:\n",
      " |          index: Value of default index. This index will be returned when OOV token is queried.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Readonly properties defined here:\n",
      " |  \n",
      " |  is_jitable\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data and other attributes defined here:\n",
      " |  \n",
      " |  __jit_unused_properties__ = ['is_jitable']\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Methods inherited from torch.nn.modules.module.Module:\n",
      " |  \n",
      " |  __call__ = _call_impl(self, *args, **kwargs)\n",
      " |  \n",
      " |  __delattr__(self, name)\n",
      " |      Implement delattr(self, name).\n",
      " |  \n",
      " |  __dir__(self)\n",
      " |      Default dir() implementation.\n",
      " |  \n",
      " |  __getattr__(self, name: str) -> Union[torch.Tensor, ForwardRef('Module')]\n",
      " |  \n",
      " |  __repr__(self)\n",
      " |      Return repr(self).\n",
      " |  \n",
      " |  __setattr__(self, name: str, value: Union[torch.Tensor, ForwardRef('Module')]) -> None\n",
      " |      Implement setattr(self, name, value).\n",
      " |  \n",
      " |  __setstate__(self, state)\n",
      " |  \n",
      " |  add_module(self, name: str, module: Union[ForwardRef('Module'), NoneType]) -> None\n",
      " |      Adds a child module to the current module.\n",
      " |      \n",
      " |      The module can be accessed as an attribute using the given name.\n",
      " |      \n",
      " |      Args:\n",
      " |          name (str): name of the child module. The child module can be\n",
      " |              accessed from this module using the given name\n",
      " |          module (Module): child module to be added to the module.\n",
      " |  \n",
      " |  apply(self: ~T, fn: Callable[[ForwardRef('Module')], NoneType]) -> ~T\n",
      " |      Applies ``fn`` recursively to every submodule (as returned by ``.children()``)\n",
      " |      as well as self. Typical use includes initializing the parameters of a model\n",
      " |      (see also :ref:`nn-init-doc`).\n",
      " |      \n",
      " |      Args:\n",
      " |          fn (:class:`Module` -> None): function to be applied to each submodule\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> @torch.no_grad()\n",
      " |          >>> def init_weights(m):\n",
      " |          >>>     print(m)\n",
      " |          >>>     if type(m) == nn.Linear:\n",
      " |          >>>         m.weight.fill_(1.0)\n",
      " |          >>>         print(m.weight)\n",
      " |          >>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))\n",
      " |          >>> net.apply(init_weights)\n",
      " |          Linear(in_features=2, out_features=2, bias=True)\n",
      " |          Parameter containing:\n",
      " |          tensor([[1., 1.],\n",
      " |                  [1., 1.]], requires_grad=True)\n",
      " |          Linear(in_features=2, out_features=2, bias=True)\n",
      " |          Parameter containing:\n",
      " |          tensor([[1., 1.],\n",
      " |                  [1., 1.]], requires_grad=True)\n",
      " |          Sequential(\n",
      " |            (0): Linear(in_features=2, out_features=2, bias=True)\n",
      " |            (1): Linear(in_features=2, out_features=2, bias=True)\n",
      " |          )\n",
      " |  \n",
      " |  bfloat16(self: ~T) -> ~T\n",
      " |      Casts all floating point parameters and buffers to ``bfloat16`` datatype.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  buffers(self, recurse: bool = True) -> Iterator[torch.Tensor]\n",
      " |      Returns an iterator over module buffers.\n",
      " |      \n",
      " |      Args:\n",
      " |          recurse (bool): if True, then yields buffers of this module\n",
      " |              and all submodules. Otherwise, yields only buffers that\n",
      " |              are direct members of this module.\n",
      " |      \n",
      " |      Yields:\n",
      " |          torch.Tensor: module buffer\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> # xdoctest: +SKIP(\"undefined vars\")\n",
      " |          >>> for buf in model.buffers():\n",
      " |          >>>     print(type(buf), buf.size())\n",
      " |          <class 'torch.Tensor'> (20L,)\n",
      " |          <class 'torch.Tensor'> (20L, 1L, 5L, 5L)\n",
      " |  \n",
      " |  children(self) -> Iterator[ForwardRef('Module')]\n",
      " |      Returns an iterator over immediate children modules.\n",
      " |      \n",
      " |      Yields:\n",
      " |          Module: a child module\n",
      " |  \n",
      " |  cpu(self: ~T) -> ~T\n",
      " |      Moves all model parameters and buffers to the CPU.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  cuda(self: ~T, device: Union[int, torch.device, NoneType] = None) -> ~T\n",
      " |      Moves all model parameters and buffers to the GPU.\n",
      " |      \n",
      " |      This also makes associated parameters and buffers different objects. So\n",
      " |      it should be called before constructing optimizer if the module will\n",
      " |      live on GPU while being optimized.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Args:\n",
      " |          device (int, optional): if specified, all parameters will be\n",
      " |              copied to that device\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  double(self: ~T) -> ~T\n",
      " |      Casts all floating point parameters and buffers to ``double`` datatype.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  eval(self: ~T) -> ~T\n",
      " |      Sets the module in evaluation mode.\n",
      " |      \n",
      " |      This has any effect only on certain modules. See documentations of\n",
      " |      particular modules for details of their behaviors in training/evaluation\n",
      " |      mode, if they are affected, e.g. :class:`Dropout`, :class:`BatchNorm`,\n",
      " |      etc.\n",
      " |      \n",
      " |      This is equivalent with :meth:`self.train(False) <torch.nn.Module.train>`.\n",
      " |      \n",
      " |      See :ref:`locally-disable-grad-doc` for a comparison between\n",
      " |      `.eval()` and several similar mechanisms that may be confused with it.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  extra_repr(self) -> str\n",
      " |      Set the extra representation of the module\n",
      " |      \n",
      " |      To print customized extra information, you should re-implement\n",
      " |      this method in your own modules. Both single-line and multi-line\n",
      " |      strings are acceptable.\n",
      " |  \n",
      " |  float(self: ~T) -> ~T\n",
      " |      Casts all floating point parameters and buffers to ``float`` datatype.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  get_buffer(self, target: str) -> 'Tensor'\n",
      " |      Returns the buffer given by ``target`` if it exists,\n",
      " |      otherwise throws an error.\n",
      " |      \n",
      " |      See the docstring for ``get_submodule`` for a more detailed\n",
      " |      explanation of this method's functionality as well as how to\n",
      " |      correctly specify ``target``.\n",
      " |      \n",
      " |      Args:\n",
      " |          target: The fully-qualified string name of the buffer\n",
      " |              to look for. (See ``get_submodule`` for how to specify a\n",
      " |              fully-qualified string.)\n",
      " |      \n",
      " |      Returns:\n",
      " |          torch.Tensor: The buffer referenced by ``target``\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: If the target string references an invalid\n",
      " |              path or resolves to something that is not a\n",
      " |              buffer\n",
      " |  \n",
      " |  get_extra_state(self) -> Any\n",
      " |      Returns any extra state to include in the module's state_dict.\n",
      " |      Implement this and a corresponding :func:`set_extra_state` for your module\n",
      " |      if you need to store extra state. This function is called when building the\n",
      " |      module's `state_dict()`.\n",
      " |      \n",
      " |      Note that extra state should be picklable to ensure working serialization\n",
      " |      of the state_dict. We only provide provide backwards compatibility guarantees\n",
      " |      for serializing Tensors; other objects may break backwards compatibility if\n",
      " |      their serialized pickled form changes.\n",
      " |      \n",
      " |      Returns:\n",
      " |          object: Any extra state to store in the module's state_dict\n",
      " |  \n",
      " |  get_parameter(self, target: str) -> 'Parameter'\n",
      " |      Returns the parameter given by ``target`` if it exists,\n",
      " |      otherwise throws an error.\n",
      " |      \n",
      " |      See the docstring for ``get_submodule`` for a more detailed\n",
      " |      explanation of this method's functionality as well as how to\n",
      " |      correctly specify ``target``.\n",
      " |      \n",
      " |      Args:\n",
      " |          target: The fully-qualified string name of the Parameter\n",
      " |              to look for. (See ``get_submodule`` for how to specify a\n",
      " |              fully-qualified string.)\n",
      " |      \n",
      " |      Returns:\n",
      " |          torch.nn.Parameter: The Parameter referenced by ``target``\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: If the target string references an invalid\n",
      " |              path or resolves to something that is not an\n",
      " |              ``nn.Parameter``\n",
      " |  \n",
      " |  get_submodule(self, target: str) -> 'Module'\n",
      " |      Returns the submodule given by ``target`` if it exists,\n",
      " |      otherwise throws an error.\n",
      " |      \n",
      " |      For example, let's say you have an ``nn.Module`` ``A`` that\n",
      " |      looks like this:\n",
      " |      \n",
      " |      .. code-block:: text\n",
      " |      \n",
      " |          A(\n",
      " |              (net_b): Module(\n",
      " |                  (net_c): Module(\n",
      " |                      (conv): Conv2d(16, 33, kernel_size=(3, 3), stride=(2, 2))\n",
      " |                  )\n",
      " |                  (linear): Linear(in_features=100, out_features=200, bias=True)\n",
      " |              )\n",
      " |          )\n",
      " |      \n",
      " |      (The diagram shows an ``nn.Module`` ``A``. ``A`` has a nested\n",
      " |      submodule ``net_b``, which itself has two submodules ``net_c``\n",
      " |      and ``linear``. ``net_c`` then has a submodule ``conv``.)\n",
      " |      \n",
      " |      To check whether or not we have the ``linear`` submodule, we\n",
      " |      would call ``get_submodule(\"net_b.linear\")``. To check whether\n",
      " |      we have the ``conv`` submodule, we would call\n",
      " |      ``get_submodule(\"net_b.net_c.conv\")``.\n",
      " |      \n",
      " |      The runtime of ``get_submodule`` is bounded by the degree\n",
      " |      of module nesting in ``target``. A query against\n",
      " |      ``named_modules`` achieves the same result, but it is O(N) in\n",
      " |      the number of transitive modules. So, for a simple check to see\n",
      " |      if some submodule exists, ``get_submodule`` should always be\n",
      " |      used.\n",
      " |      \n",
      " |      Args:\n",
      " |          target: The fully-qualified string name of the submodule\n",
      " |              to look for. (See above example for how to specify a\n",
      " |              fully-qualified string.)\n",
      " |      \n",
      " |      Returns:\n",
      " |          torch.nn.Module: The submodule referenced by ``target``\n",
      " |      \n",
      " |      Raises:\n",
      " |          AttributeError: If the target string references an invalid\n",
      " |              path or resolves to something that is not an\n",
      " |              ``nn.Module``\n",
      " |  \n",
      " |  half(self: ~T) -> ~T\n",
      " |      Casts all floating point parameters and buffers to ``half`` datatype.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  ipu(self: ~T, device: Union[int, torch.device, NoneType] = None) -> ~T\n",
      " |      Moves all model parameters and buffers to the IPU.\n",
      " |      \n",
      " |      This also makes associated parameters and buffers different objects. So\n",
      " |      it should be called before constructing optimizer if the module will\n",
      " |      live on IPU while being optimized.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Arguments:\n",
      " |          device (int, optional): if specified, all parameters will be\n",
      " |              copied to that device\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  load_state_dict(self, state_dict: Mapping[str, Any], strict: bool = True)\n",
      " |      Copies parameters and buffers from :attr:`state_dict` into\n",
      " |      this module and its descendants. If :attr:`strict` is ``True``, then\n",
      " |      the keys of :attr:`state_dict` must exactly match the keys returned\n",
      " |      by this module's :meth:`~torch.nn.Module.state_dict` function.\n",
      " |      \n",
      " |      Args:\n",
      " |          state_dict (dict): a dict containing parameters and\n",
      " |              persistent buffers.\n",
      " |          strict (bool, optional): whether to strictly enforce that the keys\n",
      " |              in :attr:`state_dict` match the keys returned by this module's\n",
      " |              :meth:`~torch.nn.Module.state_dict` function. Default: ``True``\n",
      " |      \n",
      " |      Returns:\n",
      " |          ``NamedTuple`` with ``missing_keys`` and ``unexpected_keys`` fields:\n",
      " |              * **missing_keys** is a list of str containing the missing keys\n",
      " |              * **unexpected_keys** is a list of str containing the unexpected keys\n",
      " |      \n",
      " |      Note:\n",
      " |          If a parameter or buffer is registered as ``None`` and its corresponding key\n",
      " |          exists in :attr:`state_dict`, :meth:`load_state_dict` will raise a\n",
      " |          ``RuntimeError``.\n",
      " |  \n",
      " |  modules(self) -> Iterator[ForwardRef('Module')]\n",
      " |      Returns an iterator over all modules in the network.\n",
      " |      \n",
      " |      Yields:\n",
      " |          Module: a module in the network\n",
      " |      \n",
      " |      Note:\n",
      " |          Duplicate modules are returned only once. In the following\n",
      " |          example, ``l`` will be returned only once.\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> l = nn.Linear(2, 2)\n",
      " |          >>> net = nn.Sequential(l, l)\n",
      " |          >>> for idx, m in enumerate(net.modules()):\n",
      " |          ...     print(idx, '->', m)\n",
      " |      \n",
      " |          0 -> Sequential(\n",
      " |            (0): Linear(in_features=2, out_features=2, bias=True)\n",
      " |            (1): Linear(in_features=2, out_features=2, bias=True)\n",
      " |          )\n",
      " |          1 -> Linear(in_features=2, out_features=2, bias=True)\n",
      " |  \n",
      " |  named_buffers(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> Iterator[Tuple[str, torch.Tensor]]\n",
      " |      Returns an iterator over module buffers, yielding both the\n",
      " |      name of the buffer as well as the buffer itself.\n",
      " |      \n",
      " |      Args:\n",
      " |          prefix (str): prefix to prepend to all buffer names.\n",
      " |          recurse (bool, optional): if True, then yields buffers of this module\n",
      " |              and all submodules. Otherwise, yields only buffers that\n",
      " |              are direct members of this module. Defaults to True.\n",
      " |          remove_duplicate (bool, optional): whether to remove the duplicated buffers in the result. Defaults to True.\n",
      " |      \n",
      " |      Yields:\n",
      " |          (str, torch.Tensor): Tuple containing the name and buffer\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> # xdoctest: +SKIP(\"undefined vars\")\n",
      " |          >>> for name, buf in self.named_buffers():\n",
      " |          >>>     if name in ['running_var']:\n",
      " |          >>>         print(buf.size())\n",
      " |  \n",
      " |  named_children(self) -> Iterator[Tuple[str, ForwardRef('Module')]]\n",
      " |      Returns an iterator over immediate children modules, yielding both\n",
      " |      the name of the module as well as the module itself.\n",
      " |      \n",
      " |      Yields:\n",
      " |          (str, Module): Tuple containing a name and child module\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> # xdoctest: +SKIP(\"undefined vars\")\n",
      " |          >>> for name, module in model.named_children():\n",
      " |          >>>     if name in ['conv4', 'conv5']:\n",
      " |          >>>         print(module)\n",
      " |  \n",
      " |  named_modules(self, memo: Union[Set[ForwardRef('Module')], NoneType] = None, prefix: str = '', remove_duplicate: bool = True)\n",
      " |      Returns an iterator over all modules in the network, yielding\n",
      " |      both the name of the module as well as the module itself.\n",
      " |      \n",
      " |      Args:\n",
      " |          memo: a memo to store the set of modules already added to the result\n",
      " |          prefix: a prefix that will be added to the name of the module\n",
      " |          remove_duplicate: whether to remove the duplicated module instances in the result\n",
      " |              or not\n",
      " |      \n",
      " |      Yields:\n",
      " |          (str, Module): Tuple of name and module\n",
      " |      \n",
      " |      Note:\n",
      " |          Duplicate modules are returned only once. In the following\n",
      " |          example, ``l`` will be returned only once.\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> l = nn.Linear(2, 2)\n",
      " |          >>> net = nn.Sequential(l, l)\n",
      " |          >>> for idx, m in enumerate(net.named_modules()):\n",
      " |          ...     print(idx, '->', m)\n",
      " |      \n",
      " |          0 -> ('', Sequential(\n",
      " |            (0): Linear(in_features=2, out_features=2, bias=True)\n",
      " |            (1): Linear(in_features=2, out_features=2, bias=True)\n",
      " |          ))\n",
      " |          1 -> ('0', Linear(in_features=2, out_features=2, bias=True))\n",
      " |  \n",
      " |  named_parameters(self, prefix: str = '', recurse: bool = True, remove_duplicate: bool = True) -> Iterator[Tuple[str, torch.nn.parameter.Parameter]]\n",
      " |      Returns an iterator over module parameters, yielding both the\n",
      " |      name of the parameter as well as the parameter itself.\n",
      " |      \n",
      " |      Args:\n",
      " |          prefix (str): prefix to prepend to all parameter names.\n",
      " |          recurse (bool): if True, then yields parameters of this module\n",
      " |              and all submodules. Otherwise, yields only parameters that\n",
      " |              are direct members of this module.\n",
      " |          remove_duplicate (bool, optional): whether to remove the duplicated\n",
      " |              parameters in the result. Defaults to True.\n",
      " |      \n",
      " |      Yields:\n",
      " |          (str, Parameter): Tuple containing the name and parameter\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> # xdoctest: +SKIP(\"undefined vars\")\n",
      " |          >>> for name, param in self.named_parameters():\n",
      " |          >>>     if name in ['bias']:\n",
      " |          >>>         print(param.size())\n",
      " |  \n",
      " |  parameters(self, recurse: bool = True) -> Iterator[torch.nn.parameter.Parameter]\n",
      " |      Returns an iterator over module parameters.\n",
      " |      \n",
      " |      This is typically passed to an optimizer.\n",
      " |      \n",
      " |      Args:\n",
      " |          recurse (bool): if True, then yields parameters of this module\n",
      " |              and all submodules. Otherwise, yields only parameters that\n",
      " |              are direct members of this module.\n",
      " |      \n",
      " |      Yields:\n",
      " |          Parameter: module parameter\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> # xdoctest: +SKIP(\"undefined vars\")\n",
      " |          >>> for param in model.parameters():\n",
      " |          >>>     print(type(param), param.size())\n",
      " |          <class 'torch.Tensor'> (20L,)\n",
      " |          <class 'torch.Tensor'> (20L, 1L, 5L, 5L)\n",
      " |  \n",
      " |  register_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[Tuple[torch.Tensor, ...], torch.Tensor], Union[Tuple[torch.Tensor, ...], torch.Tensor]], Union[NoneType, Tuple[torch.Tensor, ...], torch.Tensor]]) -> torch.utils.hooks.RemovableHandle\n",
      " |      Registers a backward hook on the module.\n",
      " |      \n",
      " |      This function is deprecated in favor of :meth:`~torch.nn.Module.register_full_backward_hook` and\n",
      " |      the behavior of this function will change in future versions.\n",
      " |      \n",
      " |      Returns:\n",
      " |          :class:`torch.utils.hooks.RemovableHandle`:\n",
      " |              a handle that can be used to remove the added hook by calling\n",
      " |              ``handle.remove()``\n",
      " |  \n",
      " |  register_buffer(self, name: str, tensor: Union[torch.Tensor, NoneType], persistent: bool = True) -> None\n",
      " |      Adds a buffer to the module.\n",
      " |      \n",
      " |      This is typically used to register a buffer that should not to be\n",
      " |      considered a model parameter. For example, BatchNorm's ``running_mean``\n",
      " |      is not a parameter, but is part of the module's state. Buffers, by\n",
      " |      default, are persistent and will be saved alongside parameters. This\n",
      " |      behavior can be changed by setting :attr:`persistent` to ``False``. The\n",
      " |      only difference between a persistent buffer and a non-persistent buffer\n",
      " |      is that the latter will not be a part of this module's\n",
      " |      :attr:`state_dict`.\n",
      " |      \n",
      " |      Buffers can be accessed as attributes using given names.\n",
      " |      \n",
      " |      Args:\n",
      " |          name (str): name of the buffer. The buffer can be accessed\n",
      " |              from this module using the given name\n",
      " |          tensor (Tensor or None): buffer to be registered. If ``None``, then operations\n",
      " |              that run on buffers, such as :attr:`cuda`, are ignored. If ``None``,\n",
      " |              the buffer is **not** included in the module's :attr:`state_dict`.\n",
      " |          persistent (bool): whether the buffer is part of this module's\n",
      " |              :attr:`state_dict`.\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> # xdoctest: +SKIP(\"undefined vars\")\n",
      " |          >>> self.register_buffer('running_mean', torch.zeros(num_features))\n",
      " |  \n",
      " |  register_forward_hook(self, hook: Union[Callable[[~T, Tuple[Any, ...], Any], Union[Any, NoneType]], Callable[[~T, Tuple[Any, ...], Dict[str, Any], Any], Union[Any, NoneType]]], *, prepend: bool = False, with_kwargs: bool = False) -> torch.utils.hooks.RemovableHandle\n",
      " |      Registers a forward hook on the module.\n",
      " |      \n",
      " |      The hook will be called every time after :func:`forward` has computed an output.\n",
      " |      \n",
      " |      If ``with_kwargs`` is ``False`` or not specified, the input contains only\n",
      " |      the positional arguments given to the module. Keyword arguments won't be\n",
      " |      passed to the hooks and only to the ``forward``. The hook can modify the\n",
      " |      output. It can modify the input inplace but it will not have effect on\n",
      " |      forward since this is called after :func:`forward` is called. The hook\n",
      " |      should have the following signature::\n",
      " |      \n",
      " |          hook(module, args, output) -> None or modified output\n",
      " |      \n",
      " |      If ``with_kwargs`` is ``True``, the forward hook will be passed the\n",
      " |      ``kwargs`` given to the forward function and be expected to return the\n",
      " |      output possibly modified. The hook should have the following signature::\n",
      " |      \n",
      " |          hook(module, args, kwargs, output) -> None or modified output\n",
      " |      \n",
      " |      Args:\n",
      " |          hook (Callable): The user defined hook to be registered.\n",
      " |          prepend (bool): If ``True``, the provided ``hook`` will be fired\n",
      " |              before all existing ``forward`` hooks on this\n",
      " |              :class:`torch.nn.modules.Module`. Otherwise, the provided\n",
      " |              ``hook`` will be fired after all existing ``forward`` hooks on\n",
      " |              this :class:`torch.nn.modules.Module`. Note that global\n",
      " |              ``forward`` hooks registered with\n",
      " |              :func:`register_module_forward_hook` will fire before all hooks\n",
      " |              registered by this method.\n",
      " |              Default: ``False``\n",
      " |          with_kwargs (bool): If ``True``, the ``hook`` will be passed the\n",
      " |              kwargs given to the forward function.\n",
      " |              Default: ``False``\n",
      " |      \n",
      " |      Returns:\n",
      " |          :class:`torch.utils.hooks.RemovableHandle`:\n",
      " |              a handle that can be used to remove the added hook by calling\n",
      " |              ``handle.remove()``\n",
      " |  \n",
      " |  register_forward_pre_hook(self, hook: Union[Callable[[~T, Tuple[Any, ...]], Union[Any, NoneType]], Callable[[~T, Tuple[Any, ...], Dict[str, Any]], Union[Tuple[Any, Dict[str, Any]], NoneType]]], *, prepend: bool = False, with_kwargs: bool = False) -> torch.utils.hooks.RemovableHandle\n",
      " |      Registers a forward pre-hook on the module.\n",
      " |      \n",
      " |      The hook will be called every time before :func:`forward` is invoked.\n",
      " |      \n",
      " |      \n",
      " |      If ``with_kwargs`` is false or not specified, the input contains only\n",
      " |      the positional arguments given to the module. Keyword arguments won't be\n",
      " |      passed to the hooks and only to the ``forward``. The hook can modify the\n",
      " |      input. User can either return a tuple or a single modified value in the\n",
      " |      hook. We will wrap the value into a tuple if a single value is returned\n",
      " |      (unless that value is already a tuple). The hook should have the\n",
      " |      following signature::\n",
      " |      \n",
      " |          hook(module, args) -> None or modified input\n",
      " |      \n",
      " |      If ``with_kwargs`` is true, the forward pre-hook will be passed the\n",
      " |      kwargs given to the forward function. And if the hook modifies the\n",
      " |      input, both the args and kwargs should be returned. The hook should have\n",
      " |      the following signature::\n",
      " |      \n",
      " |          hook(module, args, kwargs) -> None or a tuple of modified input and kwargs\n",
      " |      \n",
      " |      Args:\n",
      " |          hook (Callable): The user defined hook to be registered.\n",
      " |          prepend (bool): If true, the provided ``hook`` will be fired before\n",
      " |              all existing ``forward_pre`` hooks on this\n",
      " |              :class:`torch.nn.modules.Module`. Otherwise, the provided\n",
      " |              ``hook`` will be fired after all existing ``forward_pre`` hooks\n",
      " |              on this :class:`torch.nn.modules.Module`. Note that global\n",
      " |              ``forward_pre`` hooks registered with\n",
      " |              :func:`register_module_forward_pre_hook` will fire before all\n",
      " |              hooks registered by this method.\n",
      " |              Default: ``False``\n",
      " |          with_kwargs (bool): If true, the ``hook`` will be passed the kwargs\n",
      " |              given to the forward function.\n",
      " |              Default: ``False``\n",
      " |      \n",
      " |      Returns:\n",
      " |          :class:`torch.utils.hooks.RemovableHandle`:\n",
      " |              a handle that can be used to remove the added hook by calling\n",
      " |              ``handle.remove()``\n",
      " |  \n",
      " |  register_full_backward_hook(self, hook: Callable[[ForwardRef('Module'), Union[Tuple[torch.Tensor, ...], torch.Tensor], Union[Tuple[torch.Tensor, ...], torch.Tensor]], Union[NoneType, Tuple[torch.Tensor, ...], torch.Tensor]], prepend: bool = False) -> torch.utils.hooks.RemovableHandle\n",
      " |      Registers a backward hook on the module.\n",
      " |      \n",
      " |      The hook will be called every time the gradients with respect to a module\n",
      " |      are computed, i.e. the hook will execute if and only if the gradients with\n",
      " |      respect to module outputs are computed. The hook should have the following\n",
      " |      signature::\n",
      " |      \n",
      " |          hook(module, grad_input, grad_output) -> tuple(Tensor) or None\n",
      " |      \n",
      " |      The :attr:`grad_input` and :attr:`grad_output` are tuples that contain the gradients\n",
      " |      with respect to the inputs and outputs respectively. The hook should\n",
      " |      not modify its arguments, but it can optionally return a new gradient with\n",
      " |      respect to the input that will be used in place of :attr:`grad_input` in\n",
      " |      subsequent computations. :attr:`grad_input` will only correspond to the inputs given\n",
      " |      as positional arguments and all kwarg arguments are ignored. Entries\n",
      " |      in :attr:`grad_input` and :attr:`grad_output` will be ``None`` for all non-Tensor\n",
      " |      arguments.\n",
      " |      \n",
      " |      For technical reasons, when this hook is applied to a Module, its forward function will\n",
      " |      receive a view of each Tensor passed to the Module. Similarly the caller will receive a view\n",
      " |      of each Tensor returned by the Module's forward function.\n",
      " |      \n",
      " |      .. warning ::\n",
      " |          Modifying inputs or outputs inplace is not allowed when using backward hooks and\n",
      " |          will raise an error.\n",
      " |      \n",
      " |      Args:\n",
      " |          hook (Callable): The user-defined hook to be registered.\n",
      " |          prepend (bool): If true, the provided ``hook`` will be fired before\n",
      " |              all existing ``backward`` hooks on this\n",
      " |              :class:`torch.nn.modules.Module`. Otherwise, the provided\n",
      " |              ``hook`` will be fired after all existing ``backward`` hooks on\n",
      " |              this :class:`torch.nn.modules.Module`. Note that global\n",
      " |              ``backward`` hooks registered with\n",
      " |              :func:`register_module_full_backward_hook` will fire before\n",
      " |              all hooks registered by this method.\n",
      " |      \n",
      " |      Returns:\n",
      " |          :class:`torch.utils.hooks.RemovableHandle`:\n",
      " |              a handle that can be used to remove the added hook by calling\n",
      " |              ``handle.remove()``\n",
      " |  \n",
      " |  register_full_backward_pre_hook(self, hook: Callable[[ForwardRef('Module'), Union[Tuple[torch.Tensor, ...], torch.Tensor]], Union[NoneType, Tuple[torch.Tensor, ...], torch.Tensor]], prepend: bool = False) -> torch.utils.hooks.RemovableHandle\n",
      " |      Registers a backward pre-hook on the module.\n",
      " |      \n",
      " |      The hook will be called every time the gradients for the module are computed.\n",
      " |      The hook should have the following signature::\n",
      " |      \n",
      " |          hook(module, grad_output) -> Tensor or None\n",
      " |      \n",
      " |      The :attr:`grad_output` is a tuple. The hook should\n",
      " |      not modify its arguments, but it can optionally return a new gradient with\n",
      " |      respect to the output that will be used in place of :attr:`grad_output` in\n",
      " |      subsequent computations. Entries in :attr:`grad_output` will be ``None`` for\n",
      " |      all non-Tensor arguments.\n",
      " |      \n",
      " |      For technical reasons, when this hook is applied to a Module, its forward function will\n",
      " |      receive a view of each Tensor passed to the Module. Similarly the caller will receive a view\n",
      " |      of each Tensor returned by the Module's forward function.\n",
      " |      \n",
      " |      .. warning ::\n",
      " |          Modifying inputs inplace is not allowed when using backward hooks and\n",
      " |          will raise an error.\n",
      " |      \n",
      " |      Args:\n",
      " |          hook (Callable): The user-defined hook to be registered.\n",
      " |          prepend (bool): If true, the provided ``hook`` will be fired before\n",
      " |              all existing ``backward_pre`` hooks on this\n",
      " |              :class:`torch.nn.modules.Module`. Otherwise, the provided\n",
      " |              ``hook`` will be fired after all existing ``backward_pre`` hooks\n",
      " |              on this :class:`torch.nn.modules.Module`. Note that global\n",
      " |              ``backward_pre`` hooks registered with\n",
      " |              :func:`register_module_full_backward_pre_hook` will fire before\n",
      " |              all hooks registered by this method.\n",
      " |      \n",
      " |      Returns:\n",
      " |          :class:`torch.utils.hooks.RemovableHandle`:\n",
      " |              a handle that can be used to remove the added hook by calling\n",
      " |              ``handle.remove()``\n",
      " |  \n",
      " |  register_load_state_dict_post_hook(self, hook)\n",
      " |      Registers a post hook to be run after module's ``load_state_dict``\n",
      " |      is called.\n",
      " |      \n",
      " |      It should have the following signature::\n",
      " |          hook(module, incompatible_keys) -> None\n",
      " |      \n",
      " |      The ``module`` argument is the current module that this hook is registered\n",
      " |      on, and the ``incompatible_keys`` argument is a ``NamedTuple`` consisting\n",
      " |      of attributes ``missing_keys`` and ``unexpected_keys``. ``missing_keys``\n",
      " |      is a ``list`` of ``str`` containing the missing keys and\n",
      " |      ``unexpected_keys`` is a ``list`` of ``str`` containing the unexpected keys.\n",
      " |      \n",
      " |      The given incompatible_keys can be modified inplace if needed.\n",
      " |      \n",
      " |      Note that the checks performed when calling :func:`load_state_dict` with\n",
      " |      ``strict=True`` are affected by modifications the hook makes to\n",
      " |      ``missing_keys`` or ``unexpected_keys``, as expected. Additions to either\n",
      " |      set of keys will result in an error being thrown when ``strict=True``, and\n",
      " |      clearing out both missing and unexpected keys will avoid an error.\n",
      " |      \n",
      " |      Returns:\n",
      " |          :class:`torch.utils.hooks.RemovableHandle`:\n",
      " |              a handle that can be used to remove the added hook by calling\n",
      " |              ``handle.remove()``\n",
      " |  \n",
      " |  register_module(self, name: str, module: Union[ForwardRef('Module'), NoneType]) -> None\n",
      " |      Alias for :func:`add_module`.\n",
      " |  \n",
      " |  register_parameter(self, name: str, param: Union[torch.nn.parameter.Parameter, NoneType]) -> None\n",
      " |      Adds a parameter to the module.\n",
      " |      \n",
      " |      The parameter can be accessed as an attribute using given name.\n",
      " |      \n",
      " |      Args:\n",
      " |          name (str): name of the parameter. The parameter can be accessed\n",
      " |              from this module using the given name\n",
      " |          param (Parameter or None): parameter to be added to the module. If\n",
      " |              ``None``, then operations that run on parameters, such as :attr:`cuda`,\n",
      " |              are ignored. If ``None``, the parameter is **not** included in the\n",
      " |              module's :attr:`state_dict`.\n",
      " |  \n",
      " |  register_state_dict_pre_hook(self, hook)\n",
      " |      These hooks will be called with arguments: ``self``, ``prefix``,\n",
      " |      and ``keep_vars`` before calling ``state_dict`` on ``self``. The registered\n",
      " |      hooks can be used to perform pre-processing before the ``state_dict``\n",
      " |      call is made.\n",
      " |  \n",
      " |  requires_grad_(self: ~T, requires_grad: bool = True) -> ~T\n",
      " |      Change if autograd should record operations on parameters in this\n",
      " |      module.\n",
      " |      \n",
      " |      This method sets the parameters' :attr:`requires_grad` attributes\n",
      " |      in-place.\n",
      " |      \n",
      " |      This method is helpful for freezing part of the module for finetuning\n",
      " |      or training parts of a model individually (e.g., GAN training).\n",
      " |      \n",
      " |      See :ref:`locally-disable-grad-doc` for a comparison between\n",
      " |      `.requires_grad_()` and several similar mechanisms that may be confused with it.\n",
      " |      \n",
      " |      Args:\n",
      " |          requires_grad (bool): whether autograd should record operations on\n",
      " |                                parameters in this module. Default: ``True``.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  set_extra_state(self, state: Any)\n",
      " |      This function is called from :func:`load_state_dict` to handle any extra state\n",
      " |      found within the `state_dict`. Implement this function and a corresponding\n",
      " |      :func:`get_extra_state` for your module if you need to store extra state within its\n",
      " |      `state_dict`.\n",
      " |      \n",
      " |      Args:\n",
      " |          state (dict): Extra state from the `state_dict`\n",
      " |  \n",
      " |  share_memory(self: ~T) -> ~T\n",
      " |      See :meth:`torch.Tensor.share_memory_`\n",
      " |  \n",
      " |  state_dict(self, *args, destination=None, prefix='', keep_vars=False)\n",
      " |      Returns a dictionary containing references to the whole state of the module.\n",
      " |      \n",
      " |      Both parameters and persistent buffers (e.g. running averages) are\n",
      " |      included. Keys are corresponding parameter and buffer names.\n",
      " |      Parameters and buffers set to ``None`` are not included.\n",
      " |      \n",
      " |      .. note::\n",
      " |          The returned object is a shallow copy. It contains references\n",
      " |          to the module's parameters and buffers.\n",
      " |      \n",
      " |      .. warning::\n",
      " |          Currently ``state_dict()`` also accepts positional arguments for\n",
      " |          ``destination``, ``prefix`` and ``keep_vars`` in order. However,\n",
      " |          this is being deprecated and keyword arguments will be enforced in\n",
      " |          future releases.\n",
      " |      \n",
      " |      .. warning::\n",
      " |          Please avoid the use of argument ``destination`` as it is not\n",
      " |          designed for end-users.\n",
      " |      \n",
      " |      Args:\n",
      " |          destination (dict, optional): If provided, the state of module will\n",
      " |              be updated into the dict and the same object is returned.\n",
      " |              Otherwise, an ``OrderedDict`` will be created and returned.\n",
      " |              Default: ``None``.\n",
      " |          prefix (str, optional): a prefix added to parameter and buffer\n",
      " |              names to compose the keys in state_dict. Default: ``''``.\n",
      " |          keep_vars (bool, optional): by default the :class:`~torch.Tensor` s\n",
      " |              returned in the state dict are detached from autograd. If it's\n",
      " |              set to ``True``, detaching will not be performed.\n",
      " |              Default: ``False``.\n",
      " |      \n",
      " |      Returns:\n",
      " |          dict:\n",
      " |              a dictionary containing a whole state of the module\n",
      " |      \n",
      " |      Example::\n",
      " |      \n",
      " |          >>> # xdoctest: +SKIP(\"undefined vars\")\n",
      " |          >>> module.state_dict().keys()\n",
      " |          ['bias', 'weight']\n",
      " |  \n",
      " |  to(self, *args, **kwargs)\n",
      " |      Moves and/or casts the parameters and buffers.\n",
      " |      \n",
      " |      This can be called as\n",
      " |      \n",
      " |      .. function:: to(device=None, dtype=None, non_blocking=False)\n",
      " |         :noindex:\n",
      " |      \n",
      " |      .. function:: to(dtype, non_blocking=False)\n",
      " |         :noindex:\n",
      " |      \n",
      " |      .. function:: to(tensor, non_blocking=False)\n",
      " |         :noindex:\n",
      " |      \n",
      " |      .. function:: to(memory_format=torch.channels_last)\n",
      " |         :noindex:\n",
      " |      \n",
      " |      Its signature is similar to :meth:`torch.Tensor.to`, but only accepts\n",
      " |      floating point or complex :attr:`dtype`\\ s. In addition, this method will\n",
      " |      only cast the floating point or complex parameters and buffers to :attr:`dtype`\n",
      " |      (if given). The integral parameters and buffers will be moved\n",
      " |      :attr:`device`, if that is given, but with dtypes unchanged. When\n",
      " |      :attr:`non_blocking` is set, it tries to convert/move asynchronously\n",
      " |      with respect to the host if possible, e.g., moving CPU Tensors with\n",
      " |      pinned memory to CUDA devices.\n",
      " |      \n",
      " |      See below for examples.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Args:\n",
      " |          device (:class:`torch.device`): the desired device of the parameters\n",
      " |              and buffers in this module\n",
      " |          dtype (:class:`torch.dtype`): the desired floating point or complex dtype of\n",
      " |              the parameters and buffers in this module\n",
      " |          tensor (torch.Tensor): Tensor whose dtype and device are the desired\n",
      " |              dtype and device for all parameters and buffers in this module\n",
      " |          memory_format (:class:`torch.memory_format`): the desired memory\n",
      " |              format for 4D parameters and buffers in this module (keyword\n",
      " |              only argument)\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |      \n",
      " |      Examples::\n",
      " |      \n",
      " |          >>> # xdoctest: +IGNORE_WANT(\"non-deterministic\")\n",
      " |          >>> linear = nn.Linear(2, 2)\n",
      " |          >>> linear.weight\n",
      " |          Parameter containing:\n",
      " |          tensor([[ 0.1913, -0.3420],\n",
      " |                  [-0.5113, -0.2325]])\n",
      " |          >>> linear.to(torch.double)\n",
      " |          Linear(in_features=2, out_features=2, bias=True)\n",
      " |          >>> linear.weight\n",
      " |          Parameter containing:\n",
      " |          tensor([[ 0.1913, -0.3420],\n",
      " |                  [-0.5113, -0.2325]], dtype=torch.float64)\n",
      " |          >>> # xdoctest: +REQUIRES(env:TORCH_DOCTEST_CUDA1)\n",
      " |          >>> gpu1 = torch.device(\"cuda:1\")\n",
      " |          >>> linear.to(gpu1, dtype=torch.half, non_blocking=True)\n",
      " |          Linear(in_features=2, out_features=2, bias=True)\n",
      " |          >>> linear.weight\n",
      " |          Parameter containing:\n",
      " |          tensor([[ 0.1914, -0.3420],\n",
      " |                  [-0.5112, -0.2324]], dtype=torch.float16, device='cuda:1')\n",
      " |          >>> cpu = torch.device(\"cpu\")\n",
      " |          >>> linear.to(cpu)\n",
      " |          Linear(in_features=2, out_features=2, bias=True)\n",
      " |          >>> linear.weight\n",
      " |          Parameter containing:\n",
      " |          tensor([[ 0.1914, -0.3420],\n",
      " |                  [-0.5112, -0.2324]], dtype=torch.float16)\n",
      " |      \n",
      " |          >>> linear = nn.Linear(2, 2, bias=None).to(torch.cdouble)\n",
      " |          >>> linear.weight\n",
      " |          Parameter containing:\n",
      " |          tensor([[ 0.3741+0.j,  0.2382+0.j],\n",
      " |                  [ 0.5593+0.j, -0.4443+0.j]], dtype=torch.complex128)\n",
      " |          >>> linear(torch.ones(3, 2, dtype=torch.cdouble))\n",
      " |          tensor([[0.6122+0.j, 0.1150+0.j],\n",
      " |                  [0.6122+0.j, 0.1150+0.j],\n",
      " |                  [0.6122+0.j, 0.1150+0.j]], dtype=torch.complex128)\n",
      " |  \n",
      " |  to_empty(self: ~T, *, device: Union[str, torch.device]) -> ~T\n",
      " |      Moves the parameters and buffers to the specified device without copying storage.\n",
      " |      \n",
      " |      Args:\n",
      " |          device (:class:`torch.device`): The desired device of the parameters\n",
      " |              and buffers in this module.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  train(self: ~T, mode: bool = True) -> ~T\n",
      " |      Sets the module in training mode.\n",
      " |      \n",
      " |      This has any effect only on certain modules. See documentations of\n",
      " |      particular modules for details of their behaviors in training/evaluation\n",
      " |      mode, if they are affected, e.g. :class:`Dropout`, :class:`BatchNorm`,\n",
      " |      etc.\n",
      " |      \n",
      " |      Args:\n",
      " |          mode (bool): whether to set training mode (``True``) or evaluation\n",
      " |                       mode (``False``). Default: ``True``.\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  type(self: ~T, dst_type: Union[torch.dtype, str]) -> ~T\n",
      " |      Casts all parameters and buffers to :attr:`dst_type`.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Args:\n",
      " |          dst_type (type or string): the desired type\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  xpu(self: ~T, device: Union[int, torch.device, NoneType] = None) -> ~T\n",
      " |      Moves all model parameters and buffers to the XPU.\n",
      " |      \n",
      " |      This also makes associated parameters and buffers different objects. So\n",
      " |      it should be called before constructing optimizer if the module will\n",
      " |      live on XPU while being optimized.\n",
      " |      \n",
      " |      .. note::\n",
      " |          This method modifies the module in-place.\n",
      " |      \n",
      " |      Arguments:\n",
      " |          device (int, optional): if specified, all parameters will be\n",
      " |              copied to that device\n",
      " |      \n",
      " |      Returns:\n",
      " |          Module: self\n",
      " |  \n",
      " |  zero_grad(self, set_to_none: bool = True) -> None\n",
      " |      Sets gradients of all model parameters to zero. See similar function\n",
      " |      under :class:`torch.optim.Optimizer` for more context.\n",
      " |      \n",
      " |      Args:\n",
      " |          set_to_none (bool): instead of setting to zero, set the grads to None.\n",
      " |              See :meth:`torch.optim.Optimizer.zero_grad` for details.\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data descriptors inherited from torch.nn.modules.module.Module:\n",
      " |  \n",
      " |  __dict__\n",
      " |      dictionary for instance variables (if defined)\n",
      " |  \n",
      " |  __weakref__\n",
      " |      list of weak references to the object (if defined)\n",
      " |  \n",
      " |  ----------------------------------------------------------------------\n",
      " |  Data and other attributes inherited from torch.nn.modules.module.Module:\n",
      " |  \n",
      " |  T_destination = ~T_destination\n",
      " |  \n",
      " |  __annotations__ = {'__call__': typing.Callable[..., typing.Any], '_bac...\n",
      " |  \n",
      " |  call_super_init = False\n",
      " |  \n",
      " |  dump_patches = False\n",
      "\n"
     ]
    }
   ],
   "source": [
    "help(vocab)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1000\n",
      "2000\n",
      "3000\n",
      "4000\n",
      "5000\n",
      "6000\n",
      "7000\n",
      "8000\n",
      "9000\n",
      "10000\n",
      "11000\n",
      "12000\n",
      "13000\n",
      "14000\n",
      "15000\n",
      "16000\n",
      "17000\n",
      "18000\n",
      "19000\n",
      "20000\n",
      "21000\n",
      "22000\n",
      "23000\n",
      "24000\n",
      "25000\n",
      "26000\n",
      "27000\n",
      "28000\n",
      "29000\n",
      "30000\n",
      "31000\n",
      "32000\n",
      "33000\n",
      "34000\n",
      "35000\n",
      "36000\n",
      "37000\n",
      "38000\n",
      "39000\n",
      "40000\n",
      "41000\n",
      "42000\n",
      "43000\n",
      "44000\n",
      "45000\n",
      "46000\n",
      "47000\n",
      "48000\n",
      "49000\n",
      "50000\n",
      "51000\n",
      "52000\n",
      "53000\n",
      "54000\n",
      "55000\n",
      "56000\n",
      "57000\n",
      "58000\n",
      "59000\n",
      "60000\n",
      "61000\n",
      "62000\n",
      "63000\n",
      "64000\n",
      "65000\n",
      "66000\n",
      "67000\n",
      "68000\n",
      "69000\n",
      "70000\n",
      "71000\n",
      "72000\n",
      "73000\n",
      "74000\n",
      "75000\n",
      "76000\n",
      "77000\n",
      "78000\n",
      "79000\n",
      "80000\n",
      "81000\n",
      "82000\n",
      "83000\n",
      "84000\n",
      "85000\n",
      "86000\n",
      "87000\n",
      "88000\n",
      "89000\n",
      "90000\n",
      "91000\n",
      "92000\n",
      "93000\n",
      "94000\n",
      "95000\n",
      "96000\n",
      "97000\n",
      "98000\n",
      "99000\n",
      "100000\n",
      "101000\n",
      "102000\n",
      "103000\n",
      "104000\n",
      "105000\n",
      "106000\n",
      "107000\n",
      "108000\n",
      "109000\n",
      "110000\n",
      "111000\n",
      "112000\n",
      "113000\n",
      "114000\n",
      "115000\n",
      "116000\n",
      "117000\n",
      "118000\n",
      "119000\n",
      "120000\n",
      "121000\n",
      "122000\n",
      "123000\n",
      "124000\n",
      "125000\n",
      "126000\n",
      "127000\n",
      "128000\n",
      "129000\n",
      "130000\n",
      "131000\n",
      "132000\n",
      "133000\n",
      "134000\n",
      "135000\n",
      "136000\n",
      "137000\n",
      "138000\n",
      "139000\n",
      "140000\n",
      "141000\n",
      "142000\n",
      "143000\n",
      "144000\n",
      "145000\n",
      "146000\n",
      "147000\n",
      "148000\n",
      "149000\n",
      "150000\n",
      "151000\n",
      "152000\n",
      "153000\n",
      "154000\n",
      "155000\n",
      "156000\n",
      "157000\n",
      "158000\n",
      "159000\n",
      "160000\n",
      "161000\n",
      "162000\n",
      "163000\n",
      "164000\n",
      "165000\n",
      "166000\n",
      "167000\n",
      "168000\n",
      "169000\n",
      "170000\n",
      "171000\n",
      "172000\n",
      "173000\n",
      "174000\n",
      "175000\n",
      "176000\n",
      "177000\n",
      "178000\n",
      "179000\n",
      "180000\n",
      "181000\n",
      "182000\n",
      "183000\n",
      "184000\n",
      "185000\n",
      "186000\n",
      "187000\n",
      "188000\n",
      "189000\n",
      "190000\n",
      "191000\n",
      "192000\n",
      "193000\n",
      "194000\n",
      "195000\n",
      "196000\n",
      "197000\n",
      "198000\n",
      "199000\n",
      "200000\n",
      "201000\n",
      "202000\n",
      "203000\n",
      "204000\n",
      "205000\n",
      "206000\n",
      "207000\n",
      "208000\n",
      "209000\n",
      "210000\n",
      "211000\n",
      "212000\n",
      "213000\n",
      "214000\n",
      "215000\n",
      "216000\n",
      "217000\n",
      "218000\n",
      "219000\n",
      "220000\n",
      "221000\n",
      "222000\n",
      "223000\n",
      "224000\n",
      "225000\n",
      "226000\n",
      "227000\n",
      "228000\n",
      "229000\n",
      "230000\n",
      "231000\n",
      "232000\n",
      "233000\n",
      "234000\n",
      "235000\n",
      "236000\n",
      "237000\n",
      "238000\n",
      "239000\n",
      "240000\n",
      "241000\n",
      "242000\n",
      "243000\n",
      "244000\n",
      "245000\n",
      "246000\n",
      "247000\n",
      "248000\n",
      "249000\n",
      "250000\n",
      "251000\n",
      "252000\n",
      "253000\n",
      "254000\n",
      "255000\n",
      "256000\n",
      "257000\n",
      "258000\n",
      "259000\n",
      "260000\n",
      "261000\n",
      "262000\n",
      "263000\n",
      "264000\n",
      "265000\n",
      "266000\n",
      "267000\n",
      "268000\n",
      "269000\n",
      "270000\n",
      "271000\n",
      "272000\n",
      "273000\n",
      "274000\n",
      "275000\n",
      "276000\n",
      "277000\n",
      "278000\n",
      "279000\n",
      "280000\n",
      "281000\n",
      "282000\n",
      "283000\n",
      "284000\n",
      "285000\n",
      "286000\n",
      "287000\n",
      "288000\n",
      "289000\n",
      "290000\n",
      "291000\n",
      "292000\n",
      "293000\n",
      "294000\n",
      "295000\n",
      "296000\n",
      "297000\n",
      "298000\n",
      "299000\n",
      "300000\n",
      "301000\n",
      "302000\n",
      "303000\n",
      "304000\n",
      "305000\n",
      "306000\n",
      "307000\n",
      "308000\n",
      "309000\n",
      "310000\n",
      "311000\n",
      "312000\n",
      "313000\n",
      "314000\n",
      "315000\n",
      "316000\n",
      "317000\n",
      "318000\n",
      "319000\n",
      "320000\n",
      "321000\n",
      "322000\n",
      "323000\n",
      "324000\n",
      "325000\n",
      "326000\n",
      "327000\n",
      "328000\n",
      "329000\n",
      "330000\n",
      "331000\n",
      "332000\n",
      "333000\n",
      "334000\n",
      "335000\n",
      "336000\n",
      "337000\n",
      "338000\n",
      "339000\n",
      "340000\n",
      "341000\n",
      "342000\n",
      "343000\n",
      "344000\n",
      "345000\n",
      "346000\n",
      "347000\n",
      "348000\n",
      "349000\n",
      "350000\n",
      "351000\n",
      "352000\n",
      "353000\n",
      "354000\n",
      "355000\n",
      "356000\n",
      "357000\n",
      "358000\n",
      "359000\n",
      "360000\n",
      "361000\n",
      "362000\n",
      "363000\n",
      "364000\n",
      "365000\n",
      "366000\n",
      "367000\n",
      "368000\n",
      "369000\n",
      "370000\n",
      "371000\n",
      "372000\n",
      "373000\n",
      "374000\n",
      "375000\n",
      "376000\n",
      "377000\n",
      "378000\n",
      "379000\n",
      "380000\n",
      "381000\n",
      "382000\n",
      "383000\n",
      "384000\n",
      "385000\n",
      "386000\n",
      "387000\n",
      "388000\n",
      "389000\n",
      "390000\n",
      "391000\n",
      "392000\n",
      "393000\n",
      "394000\n",
      "395000\n",
      "396000\n",
      "397000\n",
      "398000\n",
      "399000\n",
      "400000\n",
      "401000\n",
      "402000\n",
      "403000\n",
      "404000\n",
      "405000\n",
      "406000\n",
      "407000\n",
      "408000\n",
      "409000\n",
      "410000\n",
      "411000\n",
      "412000\n",
      "413000\n",
      "414000\n",
      "415000\n",
      "416000\n",
      "417000\n",
      "418000\n",
      "419000\n",
      "420000\n",
      "421000\n",
      "422000\n",
      "423000\n",
      "424000\n",
      "425000\n",
      "426000\n",
      "427000\n",
      "428000\n",
      "429000\n",
      "430000\n",
      "431000\n",
      "432000\n"
     ]
    }
   ],
   "source": [
    "def look_ahead_iterator(gen):\n",
    "    prev = None\n",
    "    for item in gen:\n",
    "        if prev is not None:\n",
    "            yield (prev, item)\n",
    "        prev = item\n",
    "\n",
    "class Bigrams(IterableDataset):\n",
    "  def __init__(self, text_file, vocabulary_size):\n",
    "      self.vocab = build_vocab_from_iterator(\n",
    "         get_word_lines_from_file(text_file),\n",
    "         max_tokens = vocabulary_size,\n",
    "         specials = ['<unk>'])\n",
    "      self.vocab.set_default_index(self.vocab['<unk>'])\n",
    "      self.vocabulary_size = vocabulary_size\n",
    "      self.text_file = text_file\n",
    "\n",
    "  def __iter__(self):\n",
    "     return look_ahead_iterator(\n",
    "         (self.vocab[t] for t in itertools.chain.from_iterable(get_word_lines_from_file(self.text_file))))\n",
    "\n",
    "train_dataset = Bigrams(train_file, vocab_size)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<__main__.Bigrams object at 0x7fdd26d23940>\n"
     ]
    }
   ],
   "source": [
    "print(train_dataset)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'|===========================================================================|\\n|                  PyTorch CUDA memory summary, device ID 0                 |\\n|---------------------------------------------------------------------------|\\n|            CUDA OOMs: 1            |        cudaMalloc retries: 1         |\\n|===========================================================================|\\n|        Metric         | Cur Usage  | Peak Usage | Tot Alloc  | Tot Freed  |\\n|---------------------------------------------------------------------------|\\n| Allocated memory      | 699613 KiB |   1903 MiB |   3735 MiB |   3052 MiB |\\n|       from large pool | 699414 KiB |   1903 MiB |   3734 MiB |   3051 MiB |\\n|       from small pool |    199 KiB |      1 MiB |      1 MiB |      1 MiB |\\n|---------------------------------------------------------------------------|\\n| Active memory         | 699613 KiB |   1903 MiB |   3735 MiB |   3052 MiB |\\n|       from large pool | 699414 KiB |   1903 MiB |   3734 MiB |   3051 MiB |\\n|       from small pool |    199 KiB |      1 MiB |      1 MiB |      1 MiB |\\n|---------------------------------------------------------------------------|\\n| Requested memory      | 699611 KiB |   1903 MiB |   3735 MiB |   3052 MiB |\\n|       from large pool | 699413 KiB |   1903 MiB |   3734 MiB |   3051 MiB |\\n|       from small pool |    197 KiB |      1 MiB |      1 MiB |      1 MiB |\\n|---------------------------------------------------------------------------|\\n| GPU reserved memory   | 710656 KiB |   1918 MiB |   1918 MiB |   1224 MiB |\\n|       from large pool | 708608 KiB |   1916 MiB |   1916 MiB |   1224 MiB |\\n|       from small pool |   2048 KiB |      2 MiB |      2 MiB |      0 MiB |\\n|---------------------------------------------------------------------------|\\n| Non-releasable memory |  11043 KiB |  19364 KiB |  28939 KiB |  17896 KiB |\\n|       from large pool |   9194 KiB |  17514 KiB |  25954 KiB |  16760 KiB |\\n|       from small pool |   1849 KiB |   1950 KiB |   2985 KiB |   1136 KiB |\\n|---------------------------------------------------------------------------|\\n| Allocations           |      10    |      17    |      38    |      28    |\\n|       from large pool |       5    |       7    |      10    |       5    |\\n|       from small pool |       5    |      11    |      28    |      23    |\\n|---------------------------------------------------------------------------|\\n| Active allocs         |      10    |      17    |      38    |      28    |\\n|       from large pool |       5    |       7    |      10    |       5    |\\n|       from small pool |       5    |      11    |      28    |      23    |\\n|---------------------------------------------------------------------------|\\n| GPU reserved segments |       5    |       7    |       7    |       2    |\\n|       from large pool |       4    |       6    |       6    |       2    |\\n|       from small pool |       1    |       1    |       1    |       0    |\\n|---------------------------------------------------------------------------|\\n| Non-releasable allocs |       6    |       8    |      20    |      14    |\\n|       from large pool |       4    |       6    |       9    |       5    |\\n|       from small pool |       2    |       3    |      11    |       9    |\\n|---------------------------------------------------------------------------|\\n| Oversize allocations  |       0    |       0    |       0    |       0    |\\n|---------------------------------------------------------------------------|\\n| Oversize GPU segments |       0    |       0    |       0    |       0    |\\n|===========================================================================|\\n'"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "torch.cuda.memory_summary(device=None, abbreviated=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "os.environ[\"PYTORCH_CUDA_ALLOC_CONF\"] = \"max_split_size_mb:256\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "device = 'cuda'\n",
    "model = SimpleBigramNeuralLanguageModel(vocab_size, embed_size).to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "epoch: = 1\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/gedin/.local/lib/python3.8/site-packages/torch/nn/modules/container.py:217: UserWarning: Implicit dimension choice for softmax has been deprecated. Change the call to include dim=X as an argument.\n",
      "  input = module(input)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0 tensor(5.9599, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "1000\n",
      "100 tensor(6.1015, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "200 tensor(5.9708, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "2000\n",
      "300 tensor(6.2176, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "3000\n",
      "400 tensor(5.9401, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "4000\n",
      "500 tensor(6.2084, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "5000\n",
      "600 tensor(5.9736, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "6000\n",
      "700 tensor(6.1423, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "7000\n",
      "800 tensor(5.7344, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "8000\n",
      "900 tensor(6.0950, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "9000\n",
      "1000 tensor(5.8473, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "10000\n",
      "1100 tensor(6.0612, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "11000\n",
      "1200 tensor(6.1509, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "12000\n",
      "1300 tensor(6.0760, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "13000\n",
      "1400 tensor(6.2047, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "14000\n",
      "1500 tensor(6.1186, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "15000\n",
      "1600 tensor(5.8722, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "16000\n",
      "1700 tensor(5.8741, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "17000\n",
      "1800 tensor(5.8971, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "18000\n",
      "1900 tensor(5.8521, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "19000\n",
      "2000 tensor(5.9434, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "20000\n",
      "2100 tensor(6.0348, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "21000\n",
      "2200 tensor(5.8840, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "22000\n",
      "2300 tensor(5.8641, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "23000\n",
      "2400 tensor(5.9068, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "24000\n",
      "2500 tensor(5.9170, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "25000\n",
      "2600 tensor(5.9812, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "26000\n",
      "2700 tensor(5.8985, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "27000\n",
      "2800 tensor(6.0008, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "28000\n",
      "2900 tensor(6.1230, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "29000\n",
      "3000 tensor(5.8770, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "30000\n",
      "3100 tensor(5.9268, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "31000\n",
      "3200 tensor(5.8530, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "32000\n",
      "3300 tensor(5.8436, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "33000\n",
      "3400 tensor(5.7692, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "34000\n",
      "3500 tensor(5.8909, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "35000\n",
      "3600 tensor(5.8325, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "36000\n",
      "3700 tensor(5.8082, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "37000\n",
      "3800 tensor(5.8106, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "38000\n",
      "3900 tensor(5.6382, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "39000\n",
      "4000 tensor(5.6596, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "40000\n",
      "4100 tensor(5.9587, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "41000\n",
      "4200 tensor(5.8862, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "42000\n",
      "4300 tensor(5.9541, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "43000\n",
      "4400 tensor(5.8681, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "44000\n",
      "4500 tensor(5.6963, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "45000\n",
      "4600 tensor(6.0707, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "46000\n",
      "4700 tensor(5.7091, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "47000\n",
      "4800 tensor(5.8139, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "48000\n",
      "4900 tensor(5.8696, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "49000\n",
      "5000 tensor(5.8844, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "50000\n",
      "5100 tensor(5.9806, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "51000\n",
      "5200 tensor(6.0075, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "52000\n",
      "5300 tensor(6.0588, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "53000\n",
      "5400 tensor(5.8456, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "54000\n",
      "5500 tensor(5.9166, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "55000\n",
      "5600 tensor(5.6528, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "56000\n",
      "5700 tensor(5.8988, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "57000\n",
      "5800 tensor(5.9132, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "58000\n",
      "5900 tensor(5.9460, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "59000\n",
      "6000 tensor(5.7543, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "60000\n",
      "6100 tensor(5.8256, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "61000\n",
      "6200 tensor(5.9448, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "62000\n",
      "6300 tensor(5.7601, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "63000\n",
      "6400 tensor(5.7091, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "64000\n",
      "6500 tensor(5.5621, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "65000\n",
      "6600 tensor(5.7094, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "66000\n",
      "6700 tensor(5.6785, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "67000\n",
      "6800 tensor(5.9249, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "68000\n",
      "6900 tensor(5.8775, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "69000\n",
      "7000 tensor(5.8075, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "70000\n",
      "7100 tensor(5.5748, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "71000\n",
      "7200 tensor(5.7217, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "72000\n",
      "7300 tensor(5.9124, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "73000\n",
      "7400 tensor(5.7197, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "74000\n",
      "7500 tensor(5.6429, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "75000\n",
      "7600 tensor(5.6847, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "76000\n",
      "7700 tensor(5.7197, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "77000\n",
      "7800 tensor(5.8559, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "78000\n",
      "7900 tensor(5.5600, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "79000\n",
      "8000 tensor(5.6288, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "80000\n",
      "8100 tensor(5.7767, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "81000\n",
      "8200 tensor(5.8037, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "82000\n",
      "8300 tensor(5.7344, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "83000\n",
      "8400 tensor(5.8092, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "84000\n",
      "8500 tensor(5.8847, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "85000\n",
      "8600 tensor(5.8754, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "86000\n",
      "8700 tensor(5.9227, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "87000\n",
      "8800 tensor(5.8028, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "88000\n",
      "8900 tensor(5.6476, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "89000\n",
      "9000 tensor(5.7656, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "90000\n",
      "9100 tensor(5.7805, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "91000\n",
      "9200 tensor(5.6879, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "92000\n",
      "9300 tensor(5.7098, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "93000\n",
      "9400 tensor(5.5631, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "94000\n",
      "9500 tensor(5.6497, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "95000\n",
      "9600 tensor(5.7500, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "96000\n",
      "9700 tensor(5.6607, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "97000\n",
      "9800 tensor(5.7196, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "9900 tensor(5.5987, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "98000\n",
      "10000 tensor(5.7795, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "99000\n",
      "10100 tensor(5.6980, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "100000\n",
      "10200 tensor(5.6093, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "101000\n",
      "10300 tensor(5.6792, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "102000\n",
      "10400 tensor(5.7035, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "103000\n",
      "10500 tensor(5.8282, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "104000\n",
      "10600 tensor(5.8605, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "105000\n",
      "10700 tensor(5.7354, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "106000\n",
      "10800 tensor(5.8034, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "107000\n",
      "10900 tensor(5.6194, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "108000\n",
      "11000 tensor(5.8502, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "109000\n",
      "11100 tensor(5.4406, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "110000\n",
      "11200 tensor(5.6379, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "111000\n",
      "11300 tensor(5.6668, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "112000\n",
      "11400 tensor(5.6140, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "113000\n",
      "11500 tensor(5.6565, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "114000\n",
      "11600 tensor(5.6308, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "115000\n",
      "11700 tensor(5.5680, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "116000\n",
      "11800 tensor(5.7604, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "117000\n",
      "11900 tensor(5.5792, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "118000\n",
      "12000 tensor(5.7329, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "119000\n",
      "12100 tensor(5.7726, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "120000\n",
      "12200 tensor(5.7151, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "121000\n",
      "12300 tensor(5.8561, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "122000\n",
      "12400 tensor(5.6791, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "123000\n",
      "12500 tensor(5.5574, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "124000\n",
      "12600 tensor(5.6817, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "125000\n",
      "12700 tensor(5.5375, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "126000\n",
      "12800 tensor(5.7270, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "127000\n",
      "12900 tensor(5.6252, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "128000\n",
      "13000 tensor(5.4536, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "129000\n",
      "13100 tensor(5.6091, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "130000\n",
      "13200 tensor(5.7324, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "131000\n",
      "13300 tensor(5.5253, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "132000\n",
      "13400 tensor(5.6491, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "133000\n",
      "13500 tensor(5.5728, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "134000\n",
      "13600 tensor(5.6632, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "135000\n",
      "13700 tensor(5.6678, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "136000\n",
      "13800 tensor(5.6112, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "137000\n",
      "13900 tensor(5.4884, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "138000\n",
      "14000 tensor(5.7304, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "139000\n",
      "14100 tensor(5.4326, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "140000\n",
      "14200 tensor(5.7188, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "141000\n",
      "14300 tensor(5.6519, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "142000\n",
      "14400 tensor(5.5892, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "143000\n",
      "14500 tensor(5.7225, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "144000\n",
      "14600 tensor(5.7216, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "145000\n",
      "14700 tensor(5.5748, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "146000\n",
      "14800 tensor(6.0184, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "147000\n",
      "14900 tensor(5.6781, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "148000\n",
      "15000 tensor(5.6038, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "149000\n",
      "15100 tensor(5.7875, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "150000\n",
      "15200 tensor(5.6485, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "151000\n",
      "15300 tensor(5.5927, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "152000\n",
      "15400 tensor(5.5156, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "153000\n",
      "15500 tensor(5.6556, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "154000\n",
      "15600 tensor(5.6485, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "155000\n",
      "15700 tensor(5.5904, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "156000\n",
      "15800 tensor(5.4613, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "157000\n",
      "15900 tensor(5.6254, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "158000\n",
      "16000 tensor(5.4349, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "159000\n",
      "16100 tensor(5.5205, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "160000\n",
      "16200 tensor(5.8051, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "161000\n",
      "16300 tensor(5.6452, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "162000\n",
      "16400 tensor(5.6071, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "163000\n",
      "16500 tensor(5.7237, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "164000\n",
      "16600 tensor(5.5771, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "165000\n",
      "16700 tensor(5.5355, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "166000\n",
      "16800 tensor(5.6363, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "167000\n",
      "16900 tensor(5.3746, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "168000\n",
      "17000 tensor(5.6707, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "169000\n",
      "17100 tensor(5.5359, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "170000\n",
      "17200 tensor(5.6118, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "171000\n",
      "17300 tensor(5.6740, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "172000\n",
      "17400 tensor(5.4438, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "173000\n",
      "17500 tensor(5.5001, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "174000\n",
      "17600 tensor(5.4953, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "175000\n",
      "17700 tensor(5.5398, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "176000\n",
      "17800 tensor(5.6053, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "177000\n",
      "17900 tensor(5.4726, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "178000\n",
      "18000 tensor(5.6747, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "179000\n",
      "18100 tensor(5.6238, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "180000\n",
      "18200 tensor(5.5469, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "181000\n",
      "18300 tensor(5.5299, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "182000\n",
      "18400 tensor(5.6323, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "183000\n",
      "18500 tensor(5.5893, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "184000\n",
      "18600 tensor(5.7452, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "185000\n",
      "18700 tensor(5.5576, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "186000\n",
      "18800 tensor(5.7439, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "187000\n",
      "18900 tensor(5.6106, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "188000\n",
      "19000 tensor(5.6647, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "189000\n",
      "19100 tensor(5.7728, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "190000\n",
      "19200 tensor(5.6169, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "191000\n",
      "19300 tensor(5.7852, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "192000\n",
      "19400 tensor(5.5627, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "193000\n",
      "19500 tensor(5.5682, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "194000\n",
      "19600 tensor(5.5978, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "195000\n",
      "19700 tensor(5.6453, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "196000\n",
      "19800 tensor(5.4786, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "197000\n",
      "19900 tensor(5.4894, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "198000\n",
      "20000 tensor(5.4999, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "199000\n",
      "20100 tensor(5.4881, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "200000\n",
      "20200 tensor(5.3915, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "201000\n",
      "20300 tensor(5.5216, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "20400 tensor(5.5761, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "202000\n",
      "20500 tensor(5.5586, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "203000\n",
      "20600 tensor(5.7870, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "204000\n",
      "20700 tensor(5.5776, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "205000\n",
      "20800 tensor(5.4417, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "206000\n",
      "20900 tensor(5.7186, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "207000\n",
      "21000 tensor(5.5415, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "208000\n",
      "21100 tensor(5.5141, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "209000\n",
      "21200 tensor(5.4401, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "210000\n",
      "21300 tensor(5.6511, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "211000\n",
      "21400 tensor(5.6474, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "212000\n",
      "21500 tensor(5.3946, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "213000\n",
      "21600 tensor(5.3958, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "214000\n",
      "21700 tensor(5.4040, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "215000\n",
      "21800 tensor(5.5745, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "216000\n",
      "21900 tensor(5.4996, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "217000\n",
      "22000 tensor(5.5234, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "218000\n",
      "22100 tensor(5.3870, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "219000\n",
      "22200 tensor(5.2661, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "220000\n",
      "22300 tensor(5.7031, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "221000\n",
      "22400 tensor(5.3633, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "222000\n",
      "22500 tensor(5.4404, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "223000\n",
      "22600 tensor(5.5951, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "224000\n",
      "22700 tensor(5.3901, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "225000\n",
      "22800 tensor(5.6404, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "226000\n",
      "22900 tensor(5.6646, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "227000\n",
      "23000 tensor(5.5949, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "228000\n",
      "23100 tensor(5.5284, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "229000\n",
      "23200 tensor(5.5617, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "230000\n",
      "23300 tensor(5.6426, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "231000\n",
      "23400 tensor(5.7283, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "232000\n",
      "23500 tensor(5.4558, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "233000\n",
      "23600 tensor(5.4600, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "234000\n",
      "23700 tensor(5.4961, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "235000\n",
      "23800 tensor(5.3373, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "236000\n",
      "23900 tensor(5.4470, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "237000\n",
      "24000 tensor(5.4346, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "238000\n",
      "24100 tensor(5.5112, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "239000\n",
      "24200 tensor(5.6918, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "240000\n",
      "24300 tensor(5.6115, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "241000\n",
      "24400 tensor(5.7404, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "242000\n",
      "24500 tensor(5.4982, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "243000\n",
      "24600 tensor(5.6136, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "244000\n",
      "24700 tensor(5.5225, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "245000\n",
      "24800 tensor(5.5563, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "246000\n",
      "24900 tensor(5.6283, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "247000\n",
      "25000 tensor(5.6176, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "248000\n",
      "25100 tensor(5.5795, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "249000\n",
      "25200 tensor(5.5831, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "250000\n",
      "25300 tensor(5.5894, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "251000\n",
      "25400 tensor(5.5670, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "252000\n",
      "25500 tensor(5.5016, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "253000\n",
      "25600 tensor(5.7909, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "254000\n",
      "25700 tensor(5.5229, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "255000\n",
      "25800 tensor(5.6035, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "256000\n",
      "25900 tensor(5.5293, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "257000\n",
      "26000 tensor(5.5553, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "258000\n",
      "26100 tensor(5.4476, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "259000\n",
      "26200 tensor(5.3721, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "260000\n",
      "26300 tensor(5.6142, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "261000\n",
      "26400 tensor(5.6202, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "262000\n",
      "26500 tensor(5.3529, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "263000\n",
      "26600 tensor(5.7148, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "264000\n",
      "26700 tensor(5.5755, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "265000\n",
      "26800 tensor(5.7480, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "266000\n",
      "26900 tensor(5.5025, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "267000\n",
      "27000 tensor(5.4017, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "268000\n",
      "27100 tensor(5.3996, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "269000\n",
      "27200 tensor(5.4862, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "270000\n",
      "27300 tensor(5.6392, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "271000\n",
      "27400 tensor(5.5634, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "272000\n",
      "27500 tensor(5.4420, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "273000\n",
      "27600 tensor(5.7835, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "274000\n",
      "27700 tensor(5.5555, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "275000\n",
      "27800 tensor(5.5381, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "276000\n",
      "27900 tensor(5.6515, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "277000\n",
      "28000 tensor(5.5254, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "278000\n",
      "28100 tensor(5.4929, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "279000\n",
      "28200 tensor(5.6218, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "280000\n",
      "28300 tensor(5.2878, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "281000\n",
      "28400 tensor(5.7112, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "282000\n",
      "28500 tensor(5.5490, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "283000\n",
      "28600 tensor(5.4572, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "284000\n",
      "28700 tensor(5.6349, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "285000\n",
      "28800 tensor(5.6607, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "286000\n",
      "28900 tensor(5.5422, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "287000\n",
      "29000 tensor(5.4277, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "288000\n",
      "29100 tensor(5.1870, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "289000\n",
      "29200 tensor(5.3593, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "290000\n",
      "29300 tensor(5.6512, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "291000\n",
      "29400 tensor(5.8051, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "292000\n",
      "29500 tensor(5.5308, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "293000\n",
      "29600 tensor(5.3791, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "294000\n",
      "29700 tensor(5.6108, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "295000\n",
      "29800 tensor(5.4015, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "296000\n",
      "29900 tensor(5.6953, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "297000\n",
      "30000 tensor(5.3925, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "298000\n",
      "30100 tensor(5.4241, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "299000\n",
      "30200 tensor(5.4216, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "300000\n",
      "30300 tensor(5.5074, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "301000\n",
      "30400 tensor(5.3631, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "302000\n",
      "30500 tensor(5.5690, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "30600 tensor(5.4734, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "303000\n",
      "30700 tensor(5.5061, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "304000\n",
      "30800 tensor(5.5709, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "305000\n",
      "30900 tensor(5.5478, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "306000\n",
      "31000 tensor(5.6687, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "307000\n",
      "31100 tensor(5.2899, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "308000\n",
      "31200 tensor(5.3663, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "309000\n",
      "31300 tensor(5.6274, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "310000\n",
      "31400 tensor(5.4358, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "311000\n",
      "31500 tensor(5.5738, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "312000\n",
      "31600 tensor(5.5612, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "313000\n",
      "31700 tensor(5.5104, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "314000\n",
      "31800 tensor(5.6343, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "315000\n",
      "31900 tensor(5.2243, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "316000\n",
      "32000 tensor(5.4320, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "317000\n",
      "32100 tensor(5.3344, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "318000\n",
      "32200 tensor(5.6543, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "319000\n",
      "32300 tensor(5.6512, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "320000\n",
      "32400 tensor(5.6237, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "321000\n",
      "32500 tensor(5.4246, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "322000\n",
      "32600 tensor(5.5469, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "323000\n",
      "32700 tensor(5.5338, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "324000\n",
      "32800 tensor(5.6954, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "325000\n",
      "32900 tensor(5.5754, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "326000\n",
      "33000 tensor(5.3334, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "327000\n",
      "33100 tensor(5.5284, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "328000\n",
      "33200 tensor(5.6350, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "329000\n",
      "33300 tensor(5.4312, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "330000\n",
      "33400 tensor(5.6854, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "331000\n",
      "33500 tensor(5.4921, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "332000\n",
      "33600 tensor(5.4345, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "333000\n",
      "33700 tensor(5.4950, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "334000\n",
      "33800 tensor(5.5757, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "335000\n",
      "33900 tensor(5.3466, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "336000\n",
      "34000 tensor(5.5373, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "337000\n",
      "34100 tensor(5.5144, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "338000\n",
      "34200 tensor(5.5543, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "339000\n",
      "34300 tensor(5.3564, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "340000\n",
      "34400 tensor(5.8091, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "341000\n",
      "34500 tensor(5.6699, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "342000\n",
      "34600 tensor(5.5536, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "343000\n",
      "34700 tensor(5.6261, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "344000\n",
      "34800 tensor(5.6504, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "345000\n",
      "34900 tensor(5.7067, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "346000\n",
      "35000 tensor(5.7307, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "347000\n",
      "35100 tensor(5.4831, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "348000\n",
      "35200 tensor(5.4367, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "349000\n",
      "35300 tensor(5.6503, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "350000\n",
      "35400 tensor(5.2892, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "351000\n",
      "35500 tensor(5.4198, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "352000\n",
      "35600 tensor(5.4870, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "353000\n",
      "35700 tensor(5.4489, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "354000\n",
      "35800 tensor(5.5170, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "355000\n",
      "35900 tensor(5.4699, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "356000\n",
      "36000 tensor(5.2451, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "357000\n",
      "36100 tensor(5.6311, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "358000\n",
      "36200 tensor(5.5157, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "359000\n",
      "36300 tensor(5.7751, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "360000\n",
      "36400 tensor(5.4740, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "361000\n",
      "36500 tensor(5.4746, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "362000\n",
      "36600 tensor(5.5244, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "363000\n",
      "36700 tensor(5.3037, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "364000\n",
      "36800 tensor(5.4238, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "365000\n",
      "36900 tensor(5.5203, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "366000\n",
      "37000 tensor(5.4431, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "367000\n",
      "37100 tensor(5.4286, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "368000\n",
      "37200 tensor(5.5108, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "369000\n",
      "37300 tensor(5.4229, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "370000\n",
      "37400 tensor(5.8406, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "371000\n",
      "37500 tensor(5.4602, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "372000\n",
      "37600 tensor(5.4417, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "373000\n",
      "37700 tensor(5.6200, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "374000\n",
      "37800 tensor(5.4527, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "375000\n",
      "37900 tensor(5.4631, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "376000\n",
      "38000 tensor(5.5196, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "377000\n",
      "38100 tensor(5.5436, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "378000\n",
      "38200 tensor(5.5269, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "379000\n",
      "38300 tensor(5.4716, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "380000\n",
      "38400 tensor(5.5081, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "381000\n",
      "38500 tensor(5.5249, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "382000\n",
      "38600 tensor(5.5018, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "383000\n",
      "38700 tensor(5.4845, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "384000\n",
      "38800 tensor(5.5505, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "385000\n",
      "38900 tensor(5.6658, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "386000\n",
      "39000 tensor(5.3333, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "387000\n",
      "39100 tensor(5.5598, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "388000\n",
      "39200 tensor(5.6624, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "389000\n",
      "39300 tensor(5.4714, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "390000\n",
      "39400 tensor(5.5470, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "391000\n",
      "39500 tensor(5.6905, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "392000\n",
      "39600 tensor(5.3592, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "393000\n",
      "39700 tensor(5.3170, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "394000\n",
      "39800 tensor(5.4491, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "395000\n",
      "39900 tensor(5.2872, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "396000\n",
      "40000 tensor(5.3865, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "397000\n",
      "40100 tensor(5.4536, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "398000\n",
      "40200 tensor(5.4382, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "399000\n",
      "40300 tensor(5.4819, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "40400 tensor(5.5250, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "400000\n",
      "40500 tensor(5.4396, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "401000\n",
      "40600 tensor(5.5062, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "402000\n",
      "40700 tensor(5.5362, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "403000\n",
      "40800 tensor(5.5015, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "404000\n",
      "40900 tensor(5.4610, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "405000\n",
      "41000 tensor(5.5083, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "406000\n",
      "41100 tensor(5.4346, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "407000\n",
      "41200 tensor(5.3340, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "408000\n",
      "41300 tensor(5.4608, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "409000\n",
      "41400 tensor(5.3758, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "410000\n",
      "41500 tensor(5.5160, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "411000\n",
      "41600 tensor(5.4290, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "412000\n",
      "41700 tensor(5.4426, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "413000\n",
      "41800 tensor(5.4764, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "414000\n",
      "41900 tensor(5.4730, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "415000\n",
      "42000 tensor(5.6150, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "416000\n",
      "42100 tensor(5.3622, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "417000\n",
      "42200 tensor(5.4380, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "418000\n",
      "42300 tensor(5.5031, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "419000\n",
      "42400 tensor(5.3124, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "420000\n",
      "42500 tensor(5.4812, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "421000\n",
      "42600 tensor(5.2723, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "422000\n",
      "42700 tensor(5.5998, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "423000\n",
      "42800 tensor(5.5254, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "424000\n",
      "42900 tensor(5.3716, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "425000\n",
      "43000 tensor(5.5020, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "426000\n",
      "43100 tensor(5.5091, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "427000\n",
      "43200 tensor(5.3182, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "428000\n",
      "43300 tensor(5.4001, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "429000\n",
      "43400 tensor(5.5150, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "430000\n",
      "43500 tensor(5.2440, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "431000\n",
      "43600 tensor(5.4439, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "432000\n",
      "epoch: = 2\n",
      "0 tensor(5.3953, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "1000\n",
      "100 tensor(5.4847, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "200 tensor(5.3626, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "2000\n",
      "300 tensor(5.4127, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "3000\n",
      "400 tensor(5.3734, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "4000\n",
      "500 tensor(5.5564, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "5000\n",
      "600 tensor(5.3391, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "6000\n",
      "700 tensor(5.6198, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "7000\n",
      "800 tensor(5.2255, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "8000\n",
      "900 tensor(5.5161, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "9000\n",
      "1000 tensor(5.3517, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "10000\n",
      "1100 tensor(5.5420, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "11000\n",
      "1200 tensor(5.6031, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "12000\n",
      "1300 tensor(5.5343, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "13000\n",
      "1400 tensor(5.5547, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "14000\n",
      "1500 tensor(5.6080, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "15000\n",
      "1600 tensor(5.2940, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "16000\n",
      "1700 tensor(5.3671, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "17000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1800 tensor(5.3777, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "18000\n",
      "1900 tensor(5.3593, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "19000\n",
      "2000 tensor(5.4348, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "20000\n",
      "2100 tensor(5.5513, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "21000\n",
      "2200 tensor(5.3939, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "22000\n",
      "2300 tensor(5.4063, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "23000\n",
      "2400 tensor(5.4092, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "24000\n",
      "2500 tensor(5.4460, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "25000\n",
      "2600 tensor(5.4738, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "26000\n",
      "2700 tensor(5.4848, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "27000\n",
      "2800 tensor(5.5244, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "28000\n",
      "2900 tensor(5.6711, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "29000\n",
      "3000 tensor(5.4024, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "30000\n",
      "3100 tensor(5.4842, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "31000\n",
      "3200 tensor(5.4863, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "32000\n",
      "3300 tensor(5.4114, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "33000\n",
      "3400 tensor(5.3231, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "34000\n",
      "3500 tensor(5.4598, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "35000\n",
      "3600 tensor(5.4579, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "36000\n",
      "3700 tensor(5.3890, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "37000\n",
      "3800 tensor(5.4162, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "38000\n",
      "3900 tensor(5.2854, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "39000\n",
      "4000 tensor(5.3370, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "40000\n",
      "4100 tensor(5.5078, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "41000\n",
      "4200 tensor(5.5341, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "42000\n",
      "4300 tensor(5.4704, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "43000\n",
      "4400 tensor(5.4990, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "44000\n",
      "4500 tensor(5.3300, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "45000\n",
      "4600 tensor(5.6674, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "46000\n",
      "4700 tensor(5.3622, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "47000\n",
      "4800 tensor(5.4762, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "48000\n",
      "4900 tensor(5.5403, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "49000\n",
      "5000 tensor(5.5359, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "50000\n",
      "5100 tensor(5.6058, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "51000\n",
      "5200 tensor(5.6209, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "52000\n",
      "5300 tensor(5.6273, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "53000\n",
      "5400 tensor(5.4695, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "54000\n",
      "5500 tensor(5.5771, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "55000\n",
      "5600 tensor(5.3552, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "56000\n",
      "5700 tensor(5.5957, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "57000\n",
      "5800 tensor(5.5952, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "58000\n",
      "5900 tensor(5.5643, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "59000\n",
      "6000 tensor(5.4346, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "60000\n",
      "6100 tensor(5.4620, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "61000\n",
      "6200 tensor(5.6256, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "62000\n",
      "6300 tensor(5.4832, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "63000\n",
      "6400 tensor(5.4063, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "64000\n",
      "6500 tensor(5.2587, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "65000\n",
      "6600 tensor(5.4320, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "66000\n",
      "6700 tensor(5.3770, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "67000\n",
      "6800 tensor(5.6077, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "68000\n",
      "6900 tensor(5.5788, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "69000\n",
      "7000 tensor(5.4929, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "70000\n",
      "7100 tensor(5.2828, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "71000\n",
      "7200 tensor(5.3992, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "72000\n",
      "7300 tensor(5.6273, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "73000\n",
      "7400 tensor(5.4385, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "74000\n",
      "7500 tensor(5.3176, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "75000\n",
      "7600 tensor(5.3834, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "76000\n",
      "7700 tensor(5.4532, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "77000\n",
      "7800 tensor(5.5669, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "78000\n",
      "7900 tensor(5.2508, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "79000\n",
      "8000 tensor(5.3027, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "80000\n",
      "8100 tensor(5.4813, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "81000\n",
      "8200 tensor(5.4822, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "82000\n",
      "8300 tensor(5.4510, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "83000\n",
      "8400 tensor(5.5712, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "84000\n",
      "8500 tensor(5.5634, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "85000\n",
      "8600 tensor(5.5616, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "86000\n",
      "8700 tensor(5.6568, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "87000\n",
      "8800 tensor(5.5397, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "88000\n",
      "8900 tensor(5.3852, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "89000\n",
      "9000 tensor(5.5022, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "90000\n",
      "9100 tensor(5.5088, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "91000\n",
      "9200 tensor(5.4214, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "92000\n",
      "9300 tensor(5.4641, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "93000\n",
      "9400 tensor(5.3085, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "94000\n",
      "9500 tensor(5.3852, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "95000\n",
      "9600 tensor(5.5097, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "96000\n",
      "9700 tensor(5.4373, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "97000\n",
      "9800 tensor(5.4786, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "9900 tensor(5.3198, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "98000\n",
      "10000 tensor(5.5310, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "99000\n",
      "10100 tensor(5.4341, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "100000\n",
      "10200 tensor(5.3571, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "101000\n",
      "10300 tensor(5.4712, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "102000\n",
      "10400 tensor(5.4810, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "103000\n",
      "10500 tensor(5.5463, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "104000\n",
      "10600 tensor(5.6233, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "105000\n",
      "10700 tensor(5.4678, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "106000\n",
      "10800 tensor(5.5040, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "107000\n",
      "10900 tensor(5.3963, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "108000\n",
      "11000 tensor(5.6295, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "109000\n",
      "11100 tensor(5.2378, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "110000\n",
      "11200 tensor(5.4184, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "111000\n",
      "11300 tensor(5.4404, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "112000\n",
      "11400 tensor(5.3875, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "113000\n",
      "11500 tensor(5.4523, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "114000\n",
      "11600 tensor(5.4418, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "115000\n",
      "11700 tensor(5.3604, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "116000\n",
      "11800 tensor(5.5647, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "117000\n",
      "11900 tensor(5.3936, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "118000\n",
      "12000 tensor(5.4823, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "119000\n",
      "12100 tensor(5.5069, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "120000\n",
      "12200 tensor(5.4983, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "121000\n",
      "12300 tensor(5.6030, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "122000\n",
      "12400 tensor(5.4763, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "123000\n",
      "12500 tensor(5.3718, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "124000\n",
      "12600 tensor(5.4416, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "125000\n",
      "12700 tensor(5.3554, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "126000\n",
      "12800 tensor(5.5392, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "127000\n",
      "12900 tensor(5.4164, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "128000\n",
      "13000 tensor(5.2286, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "129000\n",
      "13100 tensor(5.4288, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "130000\n",
      "13200 tensor(5.4770, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "131000\n",
      "13300 tensor(5.3352, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "132000\n",
      "13400 tensor(5.4349, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "133000\n",
      "13500 tensor(5.3860, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "134000\n",
      "13600 tensor(5.4648, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "135000\n",
      "13700 tensor(5.4444, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "136000\n",
      "13800 tensor(5.4320, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "137000\n",
      "13900 tensor(5.2935, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "138000\n",
      "14000 tensor(5.5387, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "139000\n",
      "14100 tensor(5.2424, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "140000\n",
      "14200 tensor(5.5177, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "141000\n",
      "14300 tensor(5.4831, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "142000\n",
      "14400 tensor(5.3877, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "143000\n",
      "14500 tensor(5.4919, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "144000\n",
      "14600 tensor(5.5253, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "145000\n",
      "14700 tensor(5.3948, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "146000\n",
      "14800 tensor(5.8442, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "147000\n",
      "14900 tensor(5.4967, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "148000\n",
      "15000 tensor(5.3788, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "149000\n",
      "15100 tensor(5.5832, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "150000\n",
      "15200 tensor(5.4482, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "151000\n",
      "15300 tensor(5.4260, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "152000\n",
      "15400 tensor(5.3273, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "153000\n",
      "15500 tensor(5.4840, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "154000\n",
      "15600 tensor(5.4851, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "155000\n",
      "15700 tensor(5.3871, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "156000\n",
      "15800 tensor(5.2933, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "157000\n",
      "15900 tensor(5.4374, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "158000\n",
      "16000 tensor(5.2555, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "159000\n",
      "16100 tensor(5.3127, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "160000\n",
      "16200 tensor(5.6423, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "161000\n",
      "16300 tensor(5.4702, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "162000\n",
      "16400 tensor(5.4419, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "163000\n",
      "16500 tensor(5.5640, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "164000\n",
      "16600 tensor(5.4099, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "165000\n",
      "16700 tensor(5.3822, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "166000\n",
      "16800 tensor(5.4643, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "167000\n",
      "16900 tensor(5.2234, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "168000\n",
      "17000 tensor(5.5021, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "169000\n",
      "17100 tensor(5.3524, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "170000\n",
      "17200 tensor(5.4725, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "171000\n",
      "17300 tensor(5.5034, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "172000\n",
      "17400 tensor(5.2911, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "173000\n",
      "17500 tensor(5.3147, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "174000\n",
      "17600 tensor(5.3426, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "175000\n",
      "17700 tensor(5.3414, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "176000\n",
      "17800 tensor(5.3991, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "177000\n",
      "17900 tensor(5.2936, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "178000\n",
      "18000 tensor(5.5238, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "179000\n",
      "18100 tensor(5.4684, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "180000\n",
      "18200 tensor(5.3916, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "181000\n",
      "18300 tensor(5.3888, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "182000\n",
      "18400 tensor(5.4299, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "183000\n",
      "18500 tensor(5.4103, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "184000\n",
      "18600 tensor(5.5980, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "185000\n",
      "18700 tensor(5.4135, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "186000\n",
      "18800 tensor(5.5855, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "187000\n",
      "18900 tensor(5.4583, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "188000\n",
      "19000 tensor(5.4854, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "189000\n",
      "19100 tensor(5.5879, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "190000\n",
      "19200 tensor(5.4675, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "191000\n",
      "19300 tensor(5.5741, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "192000\n",
      "19400 tensor(5.3977, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "193000\n",
      "19500 tensor(5.4042, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "194000\n",
      "19600 tensor(5.4364, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "195000\n",
      "19700 tensor(5.4868, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "196000\n",
      "19800 tensor(5.3476, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "197000\n",
      "19900 tensor(5.3553, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "198000\n",
      "20000 tensor(5.3707, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "199000\n",
      "20100 tensor(5.3226, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "200000\n",
      "20200 tensor(5.2488, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "201000\n",
      "20300 tensor(5.3648, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "20400 tensor(5.4156, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "202000\n",
      "20500 tensor(5.4102, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "203000\n",
      "20600 tensor(5.6109, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "204000\n",
      "20700 tensor(5.4335, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "205000\n",
      "20800 tensor(5.2795, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "206000\n",
      "20900 tensor(5.5609, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "207000\n",
      "21000 tensor(5.3918, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "208000\n",
      "21100 tensor(5.3831, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "209000\n",
      "21200 tensor(5.2790, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "210000\n",
      "21300 tensor(5.4710, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "211000\n",
      "21400 tensor(5.5050, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "212000\n",
      "21500 tensor(5.2692, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "213000\n",
      "21600 tensor(5.2668, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "214000\n",
      "21700 tensor(5.2633, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "215000\n",
      "21800 tensor(5.4067, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "216000\n",
      "21900 tensor(5.3829, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "217000\n",
      "22000 tensor(5.3773, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "218000\n",
      "22100 tensor(5.2472, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "219000\n",
      "22200 tensor(5.1171, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "220000\n",
      "22300 tensor(5.5545, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "221000\n",
      "22400 tensor(5.2499, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "222000\n",
      "22500 tensor(5.2943, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "223000\n",
      "22600 tensor(5.4748, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "224000\n",
      "22700 tensor(5.2436, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "225000\n",
      "22800 tensor(5.5053, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "226000\n",
      "22900 tensor(5.5519, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "227000\n",
      "23000 tensor(5.4541, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "228000\n",
      "23100 tensor(5.4279, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "229000\n",
      "23200 tensor(5.4286, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "230000\n",
      "23300 tensor(5.5179, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "231000\n",
      "23400 tensor(5.5355, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "232000\n",
      "23500 tensor(5.3505, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "233000\n",
      "23600 tensor(5.3313, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "234000\n",
      "23700 tensor(5.3509, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "235000\n",
      "23800 tensor(5.2170, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "236000\n",
      "23900 tensor(5.3101, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "237000\n",
      "24000 tensor(5.2962, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "238000\n",
      "24100 tensor(5.3882, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "239000\n",
      "24200 tensor(5.5633, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "240000\n",
      "24300 tensor(5.4595, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "241000\n",
      "24400 tensor(5.5932, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "242000\n",
      "24500 tensor(5.3717, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "243000\n",
      "24600 tensor(5.4943, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "244000\n",
      "24700 tensor(5.3985, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "245000\n",
      "24800 tensor(5.4347, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "246000\n",
      "24900 tensor(5.5008, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "247000\n",
      "25000 tensor(5.5100, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "248000\n",
      "25100 tensor(5.4427, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "249000\n",
      "25200 tensor(5.4508, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "250000\n",
      "25300 tensor(5.4724, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "251000\n",
      "25400 tensor(5.4525, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "252000\n",
      "25500 tensor(5.3620, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "253000\n",
      "25600 tensor(5.6446, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "254000\n",
      "25700 tensor(5.3966, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "255000\n",
      "25800 tensor(5.4889, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "256000\n",
      "25900 tensor(5.4251, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "257000\n",
      "26000 tensor(5.4346, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "258000\n",
      "26100 tensor(5.3395, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "259000\n",
      "26200 tensor(5.2695, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "260000\n",
      "26300 tensor(5.4767, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "261000\n",
      "26400 tensor(5.5083, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "262000\n",
      "26500 tensor(5.2347, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "263000\n",
      "26600 tensor(5.5761, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "264000\n",
      "26700 tensor(5.4402, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "265000\n",
      "26800 tensor(5.6173, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "266000\n",
      "26900 tensor(5.3775, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "267000\n",
      "27000 tensor(5.2863, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "268000\n",
      "27100 tensor(5.3007, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "269000\n",
      "27200 tensor(5.3551, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "270000\n",
      "27300 tensor(5.5439, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "271000\n",
      "27400 tensor(5.4334, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "272000\n",
      "27500 tensor(5.3266, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "273000\n",
      "27600 tensor(5.6412, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "274000\n",
      "27700 tensor(5.4420, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "275000\n",
      "27800 tensor(5.4381, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "276000\n",
      "27900 tensor(5.5550, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "277000\n",
      "28000 tensor(5.4154, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "278000\n",
      "28100 tensor(5.3823, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "279000\n",
      "28200 tensor(5.5344, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "280000\n",
      "28300 tensor(5.1615, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "281000\n",
      "28400 tensor(5.6069, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "282000\n",
      "28500 tensor(5.4426, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "283000\n",
      "28600 tensor(5.3672, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "284000\n",
      "28700 tensor(5.5133, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "285000\n",
      "28800 tensor(5.5556, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "286000\n",
      "28900 tensor(5.4294, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "287000\n",
      "29000 tensor(5.3359, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "288000\n",
      "29100 tensor(5.0951, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "289000\n",
      "29200 tensor(5.2511, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "290000\n",
      "29300 tensor(5.5364, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "291000\n",
      "29400 tensor(5.6708, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "292000\n",
      "29500 tensor(5.4371, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "293000\n",
      "29600 tensor(5.2942, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "294000\n",
      "29700 tensor(5.4637, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "295000\n",
      "29800 tensor(5.2914, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "296000\n",
      "29900 tensor(5.5562, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "297000\n",
      "30000 tensor(5.2833, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "298000\n",
      "30100 tensor(5.3481, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "299000\n",
      "30200 tensor(5.3122, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "300000\n",
      "30300 tensor(5.4103, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "301000\n",
      "30400 tensor(5.2480, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "302000\n",
      "30500 tensor(5.4258, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "30600 tensor(5.3835, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "303000\n",
      "30700 tensor(5.4193, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "304000\n",
      "30800 tensor(5.4438, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "305000\n",
      "30900 tensor(5.4518, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "306000\n",
      "31000 tensor(5.5607, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "307000\n",
      "31100 tensor(5.2059, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "308000\n",
      "31200 tensor(5.2571, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "309000\n",
      "31300 tensor(5.5208, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "310000\n",
      "31400 tensor(5.3061, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "311000\n",
      "31500 tensor(5.4834, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "312000\n",
      "31600 tensor(5.4653, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "313000\n",
      "31700 tensor(5.4308, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "314000\n",
      "31800 tensor(5.5400, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "315000\n",
      "31900 tensor(5.1536, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "316000\n",
      "32000 tensor(5.3460, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "317000\n",
      "32100 tensor(5.2300, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "318000\n",
      "32200 tensor(5.5511, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "319000\n",
      "32300 tensor(5.5391, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "320000\n",
      "32400 tensor(5.5157, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "321000\n",
      "32500 tensor(5.3336, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "322000\n",
      "32600 tensor(5.4475, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "323000\n",
      "32700 tensor(5.3894, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "324000\n",
      "32800 tensor(5.6022, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "325000\n",
      "32900 tensor(5.4663, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "326000\n",
      "33000 tensor(5.2387, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "327000\n",
      "33100 tensor(5.4446, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "328000\n",
      "33200 tensor(5.5450, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "329000\n",
      "33300 tensor(5.3179, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "330000\n",
      "33400 tensor(5.5905, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "331000\n",
      "33500 tensor(5.4066, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "332000\n",
      "33600 tensor(5.3542, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "333000\n",
      "33700 tensor(5.4097, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "334000\n",
      "33800 tensor(5.4912, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "335000\n",
      "33900 tensor(5.2358, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "336000\n",
      "34000 tensor(5.4470, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "337000\n",
      "34100 tensor(5.4207, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "338000\n",
      "34200 tensor(5.4651, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "339000\n",
      "34300 tensor(5.2545, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "340000\n",
      "34400 tensor(5.7106, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "341000\n",
      "34500 tensor(5.5699, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "342000\n",
      "34600 tensor(5.4638, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "343000\n",
      "34700 tensor(5.5382, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "344000\n",
      "34800 tensor(5.5603, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "345000\n",
      "34900 tensor(5.6072, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "346000\n",
      "35000 tensor(5.6037, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "347000\n",
      "35100 tensor(5.4069, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "348000\n",
      "35200 tensor(5.3398, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "349000\n",
      "35300 tensor(5.5607, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "350000\n",
      "35400 tensor(5.2068, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "351000\n",
      "35500 tensor(5.3112, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "352000\n",
      "35600 tensor(5.4126, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "353000\n",
      "35700 tensor(5.3091, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "354000\n",
      "35800 tensor(5.4252, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "355000\n",
      "35900 tensor(5.3956, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "356000\n",
      "36000 tensor(5.1705, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "357000\n",
      "36100 tensor(5.5497, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "358000\n",
      "36200 tensor(5.4066, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "359000\n",
      "36300 tensor(5.6858, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "360000\n",
      "36400 tensor(5.3812, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "361000\n",
      "36500 tensor(5.3990, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "362000\n",
      "36600 tensor(5.4302, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "363000\n",
      "36700 tensor(5.2253, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "364000\n",
      "36800 tensor(5.3347, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "365000\n",
      "36900 tensor(5.4426, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "366000\n",
      "37000 tensor(5.3419, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "367000\n",
      "37100 tensor(5.3579, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "368000\n",
      "37200 tensor(5.4332, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "369000\n",
      "37300 tensor(5.3362, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "370000\n",
      "37400 tensor(5.7100, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "371000\n",
      "37500 tensor(5.3742, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "372000\n",
      "37600 tensor(5.3615, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "373000\n",
      "37700 tensor(5.5402, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "374000\n",
      "37800 tensor(5.3734, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "375000\n",
      "37900 tensor(5.3621, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "376000\n",
      "38000 tensor(5.4380, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "377000\n",
      "38100 tensor(5.4513, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "378000\n",
      "38200 tensor(5.4554, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "379000\n",
      "38300 tensor(5.3735, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "380000\n",
      "38400 tensor(5.4297, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "381000\n",
      "38500 tensor(5.4561, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "382000\n",
      "38600 tensor(5.4118, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "383000\n",
      "38700 tensor(5.3996, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "384000\n",
      "38800 tensor(5.4825, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "385000\n",
      "38900 tensor(5.5692, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "386000\n",
      "39000 tensor(5.2573, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "387000\n",
      "39100 tensor(5.4847, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "388000\n",
      "39200 tensor(5.5802, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "389000\n",
      "39300 tensor(5.3968, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "390000\n",
      "39400 tensor(5.4666, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "391000\n",
      "39500 tensor(5.5847, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "392000\n",
      "39600 tensor(5.2648, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "393000\n",
      "39700 tensor(5.2423, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "394000\n",
      "39800 tensor(5.3731, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "395000\n",
      "39900 tensor(5.2014, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "396000\n",
      "40000 tensor(5.2903, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "397000\n",
      "40100 tensor(5.3712, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "398000\n",
      "40200 tensor(5.3557, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "399000\n",
      "40300 tensor(5.4151, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "40400 tensor(5.4358, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "400000\n",
      "40500 tensor(5.3498, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "401000\n",
      "40600 tensor(5.4152, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "402000\n",
      "40700 tensor(5.4551, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "403000\n",
      "40800 tensor(5.4138, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "404000\n",
      "40900 tensor(5.3628, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "405000\n",
      "41000 tensor(5.4124, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "406000\n",
      "41100 tensor(5.3750, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "407000\n",
      "41200 tensor(5.2687, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "408000\n",
      "41300 tensor(5.3987, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "409000\n",
      "41400 tensor(5.2976, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "410000\n",
      "41500 tensor(5.4418, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "411000\n",
      "41600 tensor(5.3558, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "412000\n",
      "41700 tensor(5.3767, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "413000\n",
      "41800 tensor(5.3836, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "414000\n",
      "41900 tensor(5.3904, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "415000\n",
      "42000 tensor(5.5445, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "416000\n",
      "42100 tensor(5.2890, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "417000\n",
      "42200 tensor(5.3691, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "418000\n",
      "42300 tensor(5.4364, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "419000\n",
      "42400 tensor(5.2507, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "420000\n",
      "42500 tensor(5.4215, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "421000\n",
      "42600 tensor(5.2136, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "422000\n",
      "42700 tensor(5.5296, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "423000\n",
      "42800 tensor(5.4544, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "424000\n",
      "42900 tensor(5.3009, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "425000\n",
      "43000 tensor(5.4403, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "426000\n",
      "43100 tensor(5.4384, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "427000\n",
      "43200 tensor(5.2520, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "428000\n",
      "43300 tensor(5.2945, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "429000\n",
      "43400 tensor(5.4455, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "430000\n",
      "43500 tensor(5.1633, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "431000\n",
      "43600 tensor(5.3649, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "432000\n",
      "epoch: = 3\n",
      "0 tensor(5.3427, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "1000\n",
      "100 tensor(5.4180, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "200 tensor(5.2939, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "2000\n",
      "300 tensor(5.3083, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "3000\n",
      "400 tensor(5.3086, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "4000\n",
      "500 tensor(5.4733, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "5000\n",
      "600 tensor(5.2627, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "6000\n",
      "700 tensor(5.5664, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "7000\n",
      "800 tensor(5.1641, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "8000\n",
      "900 tensor(5.4272, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "9000\n",
      "1000 tensor(5.2926, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "10000\n",
      "1100 tensor(5.4848, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "11000\n",
      "1200 tensor(5.5283, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "12000\n",
      "1300 tensor(5.4635, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "13000\n",
      "1400 tensor(5.4590, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "14000\n",
      "1500 tensor(5.5386, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "15000\n",
      "1600 tensor(5.2150, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "16000\n",
      "1700 tensor(5.3116, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "17000\n",
      "1800 tensor(5.3130, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "18000\n",
      "1900 tensor(5.2889, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "19000\n",
      "2000 tensor(5.3574, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "20000\n",
      "2100 tensor(5.4860, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "21000\n",
      "2200 tensor(5.3206, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "22000\n",
      "2300 tensor(5.3447, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "23000\n",
      "2400 tensor(5.3333, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "24000\n",
      "2500 tensor(5.3822, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "25000\n",
      "2600 tensor(5.4039, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "26000\n",
      "2700 tensor(5.4280, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "27000\n",
      "2800 tensor(5.4575, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "28000\n",
      "2900 tensor(5.5878, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "29000\n",
      "3000 tensor(5.3311, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "30000\n",
      "3100 tensor(5.4103, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "31000\n",
      "3200 tensor(5.4323, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "32000\n",
      "3300 tensor(5.3521, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "33000\n",
      "3400 tensor(5.2512, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "34000\n",
      "3500 tensor(5.3813, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "35000\n",
      "3600 tensor(5.4000, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "36000\n",
      "3700 tensor(5.3312, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "37000\n",
      "3800 tensor(5.3553, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "38000\n",
      "3900 tensor(5.2275, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "39000\n",
      "4000 tensor(5.2883, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "40000\n",
      "4100 tensor(5.4294, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "41000\n",
      "4200 tensor(5.4801, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "42000\n",
      "4300 tensor(5.3863, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "43000\n",
      "4400 tensor(5.4470, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "44000\n",
      "4500 tensor(5.2610, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "45000\n",
      "4600 tensor(5.5962, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "46000\n",
      "4700 tensor(5.3029, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "47000\n",
      "4800 tensor(5.4265, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "48000\n",
      "4900 tensor(5.4823, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "49000\n",
      "5000 tensor(5.4749, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "50000\n",
      "5100 tensor(5.5356, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "51000\n",
      "5200 tensor(5.5513, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "52000\n",
      "5300 tensor(5.5476, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "53000\n",
      "5400 tensor(5.4039, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "54000\n",
      "5500 tensor(5.5156, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "55000\n",
      "5600 tensor(5.2975, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "56000\n",
      "5700 tensor(5.5492, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "57000\n",
      "5800 tensor(5.5379, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "58000\n",
      "5900 tensor(5.4874, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "59000\n",
      "6000 tensor(5.3808, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "60000\n",
      "6100 tensor(5.3932, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "61000\n",
      "6200 tensor(5.5657, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "62000\n",
      "6300 tensor(5.4233, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "63000\n",
      "6400 tensor(5.3438, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "64000\n",
      "6500 tensor(5.2002, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "65000\n",
      "6600 tensor(5.3774, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "66000\n",
      "6700 tensor(5.3193, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "67000\n",
      "6800 tensor(5.5394, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "68000\n",
      "6900 tensor(5.5196, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "69000\n",
      "7000 tensor(5.4282, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "70000\n",
      "7100 tensor(5.2296, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "71000\n",
      "7200 tensor(5.3175, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "72000\n",
      "7300 tensor(5.5642, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "73000\n",
      "7400 tensor(5.3784, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "74000\n",
      "7500 tensor(5.2475, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "75000\n",
      "7600 tensor(5.3194, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "76000\n",
      "7700 tensor(5.3934, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "77000\n",
      "7800 tensor(5.5041, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "78000\n",
      "7900 tensor(5.1814, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "79000\n",
      "8000 tensor(5.2426, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "80000\n",
      "8100 tensor(5.4104, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "81000\n",
      "8200 tensor(5.4198, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "82000\n",
      "8300 tensor(5.3854, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "83000\n",
      "8400 tensor(5.5128, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "84000\n",
      "8500 tensor(5.4898, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "85000\n",
      "8600 tensor(5.4943, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "86000\n",
      "8700 tensor(5.6012, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "87000\n",
      "8800 tensor(5.4790, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "88000\n",
      "8900 tensor(5.3312, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "89000\n",
      "9000 tensor(5.4456, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "90000\n",
      "9100 tensor(5.4537, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "91000\n",
      "9200 tensor(5.3643, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "92000\n",
      "9300 tensor(5.4085, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "93000\n",
      "9400 tensor(5.2527, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "94000\n",
      "9500 tensor(5.3289, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "95000\n",
      "9600 tensor(5.4516, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "96000\n",
      "9700 tensor(5.3881, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "97000\n",
      "9800 tensor(5.4321, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "9900 tensor(5.2532, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "98000\n",
      "10000 tensor(5.4727, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "99000\n",
      "10100 tensor(5.3607, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "100000\n",
      "10200 tensor(5.2989, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "101000\n",
      "10300 tensor(5.4168, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "102000\n",
      "10400 tensor(5.4272, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "103000\n",
      "10500 tensor(5.4838, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "104000\n",
      "10600 tensor(5.5675, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "105000\n",
      "10700 tensor(5.4027, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "106000\n",
      "10800 tensor(5.4252, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "107000\n",
      "10900 tensor(5.3408, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "108000\n",
      "11000 tensor(5.5754, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "109000\n",
      "11100 tensor(5.1920, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "110000\n",
      "11200 tensor(5.3604, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "111000\n",
      "11300 tensor(5.3836, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "112000\n",
      "11400 tensor(5.3330, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "113000\n",
      "11500 tensor(5.4023, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "114000\n",
      "11600 tensor(5.3923, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "115000\n",
      "11700 tensor(5.3145, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "116000\n",
      "11800 tensor(5.5174, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "117000\n",
      "11900 tensor(5.3522, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "118000\n",
      "12000 tensor(5.4232, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "119000\n",
      "12100 tensor(5.4382, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "120000\n",
      "12200 tensor(5.4488, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "121000\n",
      "12300 tensor(5.5409, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "122000\n",
      "12400 tensor(5.4200, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "123000\n",
      "12500 tensor(5.3292, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "124000\n",
      "12600 tensor(5.3788, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "125000\n",
      "12700 tensor(5.3116, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "126000\n",
      "12800 tensor(5.4948, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "127000\n",
      "12900 tensor(5.3557, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "128000\n",
      "13000 tensor(5.1732, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "129000\n",
      "13100 tensor(5.3782, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "130000\n",
      "13200 tensor(5.4178, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "131000\n",
      "13300 tensor(5.2929, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "132000\n",
      "13400 tensor(5.3806, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "133000\n",
      "13500 tensor(5.3394, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "134000\n",
      "13600 tensor(5.4191, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "135000\n",
      "13700 tensor(5.3856, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "136000\n",
      "13800 tensor(5.3839, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "137000\n",
      "13900 tensor(5.2391, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "138000\n",
      "14000 tensor(5.4865, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "139000\n",
      "14100 tensor(5.1952, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "140000\n",
      "14200 tensor(5.4670, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "141000\n",
      "14300 tensor(5.4385, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "142000\n",
      "14400 tensor(5.3347, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "143000\n",
      "14500 tensor(5.4370, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "144000\n",
      "14600 tensor(5.4695, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "145000\n",
      "14700 tensor(5.3453, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "146000\n",
      "14800 tensor(5.7928, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "147000\n",
      "14900 tensor(5.4451, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "148000\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15000 tensor(5.3087, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "149000\n",
      "15100 tensor(5.5241, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "150000\n",
      "15200 tensor(5.3894, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "151000\n",
      "15300 tensor(5.3809, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "152000\n",
      "15400 tensor(5.2696, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "153000\n",
      "15500 tensor(5.4343, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "154000\n",
      "15600 tensor(5.4322, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "155000\n",
      "15700 tensor(5.3296, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "156000\n",
      "15800 tensor(5.2456, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "157000\n",
      "15900 tensor(5.3806, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "158000\n",
      "16000 tensor(5.2008, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "159000\n",
      "16100 tensor(5.2489, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "160000\n",
      "16200 tensor(5.5902, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "161000\n",
      "16300 tensor(5.4159, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "162000\n",
      "16400 tensor(5.3966, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "163000\n",
      "16500 tensor(5.5113, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "164000\n",
      "16600 tensor(5.3599, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "165000\n",
      "16700 tensor(5.3372, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "166000\n",
      "16800 tensor(5.4158, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "167000\n",
      "16900 tensor(5.1788, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "168000\n",
      "17000 tensor(5.4497, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "169000\n",
      "17100 tensor(5.2981, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "170000\n",
      "17200 tensor(5.4330, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "171000\n",
      "17300 tensor(5.4495, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "172000\n",
      "17400 tensor(5.2431, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "173000\n",
      "17500 tensor(5.2652, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "174000\n",
      "17600 tensor(5.3007, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "175000\n",
      "17700 tensor(5.2852, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "176000\n",
      "17800 tensor(5.3431, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "177000\n",
      "17900 tensor(5.2395, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "178000\n",
      "18000 tensor(5.4841, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "179000\n",
      "18100 tensor(5.4218, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "180000\n",
      "18200 tensor(5.3397, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "181000\n",
      "18300 tensor(5.3426, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "182000\n",
      "18400 tensor(5.3654, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "183000\n",
      "18500 tensor(5.3484, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "184000\n",
      "18600 tensor(5.5509, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "185000\n",
      "18700 tensor(5.3702, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "186000\n",
      "18800 tensor(5.5361, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "187000\n",
      "18900 tensor(5.4132, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "188000\n",
      "19000 tensor(5.4235, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "189000\n",
      "19100 tensor(5.5318, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "190000\n",
      "19200 tensor(5.4136, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "191000\n",
      "19300 tensor(5.5053, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "192000\n",
      "19400 tensor(5.3472, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "193000\n",
      "19500 tensor(5.3511, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "194000\n",
      "19600 tensor(5.3861, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "195000\n",
      "19700 tensor(5.4345, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "196000\n",
      "19800 tensor(5.3067, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "197000\n",
      "19900 tensor(5.3079, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "198000\n",
      "20000 tensor(5.3268, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "199000\n",
      "20100 tensor(5.2668, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "200000\n",
      "20200 tensor(5.1998, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "201000\n",
      "20300 tensor(5.3105, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "20400 tensor(5.3584, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "202000\n",
      "20500 tensor(5.3580, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "203000\n",
      "20600 tensor(5.5528, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "204000\n",
      "20700 tensor(5.3871, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "205000\n",
      "20800 tensor(5.2208, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "206000\n",
      "20900 tensor(5.5007, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "207000\n",
      "21000 tensor(5.3396, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "208000\n",
      "21100 tensor(5.3407, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "209000\n",
      "21200 tensor(5.2243, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "210000\n",
      "21300 tensor(5.4206, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "211000\n",
      "21400 tensor(5.4574, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "212000\n",
      "21500 tensor(5.2328, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "213000\n",
      "21600 tensor(5.2233, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "214000\n",
      "21700 tensor(5.2152, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "215000\n",
      "21800 tensor(5.3497, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "216000\n",
      "21900 tensor(5.3425, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "217000\n",
      "22000 tensor(5.3277, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "218000\n",
      "22100 tensor(5.2012, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "219000\n",
      "22200 tensor(5.0736, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "220000\n",
      "22300 tensor(5.5070, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "221000\n",
      "22400 tensor(5.2190, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "222000\n",
      "22500 tensor(5.2434, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "223000\n",
      "22600 tensor(5.4325, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "224000\n",
      "22700 tensor(5.1909, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "225000\n",
      "22800 tensor(5.4576, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "226000\n",
      "22900 tensor(5.5069, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "227000\n",
      "23000 tensor(5.4041, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "228000\n",
      "23100 tensor(5.3908, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "229000\n",
      "23200 tensor(5.3866, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "230000\n",
      "23300 tensor(5.4714, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "231000\n",
      "23400 tensor(5.4781, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "232000\n",
      "23500 tensor(5.3154, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "233000\n",
      "23600 tensor(5.2854, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "234000\n",
      "23700 tensor(5.3050, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "235000\n",
      "23800 tensor(5.1721, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "236000\n",
      "23900 tensor(5.2637, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "237000\n",
      "24000 tensor(5.2519, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "238000\n",
      "24100 tensor(5.3407, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "239000\n",
      "24200 tensor(5.5137, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "240000\n",
      "24300 tensor(5.4080, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "241000\n",
      "24400 tensor(5.5379, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "242000\n",
      "24500 tensor(5.3255, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "243000\n",
      "24600 tensor(5.4515, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "244000\n",
      "24700 tensor(5.3535, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "245000\n",
      "24800 tensor(5.3935, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "246000\n",
      "24900 tensor(5.4553, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "247000\n",
      "25000 tensor(5.4708, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "248000\n",
      "25100 tensor(5.3920, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "249000\n",
      "25200 tensor(5.4083, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "250000\n",
      "25300 tensor(5.4332, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "251000\n",
      "25400 tensor(5.4136, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "252000\n",
      "25500 tensor(5.3147, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "253000\n",
      "25600 tensor(5.5860, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "254000\n",
      "25700 tensor(5.3490, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "255000\n",
      "25800 tensor(5.4464, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "256000\n",
      "25900 tensor(5.3857, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "257000\n",
      "26000 tensor(5.3893, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "258000\n",
      "26100 tensor(5.3041, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "259000\n",
      "26200 tensor(5.2321, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "260000\n",
      "26300 tensor(5.4289, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "261000\n",
      "26400 tensor(5.4663, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "262000\n",
      "26500 tensor(5.1922, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "263000\n",
      "26600 tensor(5.5283, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "264000\n",
      "26700 tensor(5.3933, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "265000\n",
      "26800 tensor(5.5680, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "266000\n",
      "26900 tensor(5.3281, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "267000\n",
      "27000 tensor(5.2408, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "268000\n",
      "27100 tensor(5.2671, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "269000\n",
      "27200 tensor(5.3099, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "270000\n",
      "27300 tensor(5.5049, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "271000\n",
      "27400 tensor(5.3850, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "272000\n",
      "27500 tensor(5.2843, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "273000\n",
      "27600 tensor(5.5777, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "274000\n",
      "27700 tensor(5.4017, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "275000\n",
      "27800 tensor(5.3994, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "276000\n",
      "27900 tensor(5.5128, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "277000\n",
      "28000 tensor(5.3708, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "278000\n",
      "28100 tensor(5.3382, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "279000\n",
      "28200 tensor(5.4996, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "280000\n",
      "28300 tensor(5.1214, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "281000\n",
      "28400 tensor(5.5647, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "282000\n",
      "28500 tensor(5.3959, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "283000\n",
      "28600 tensor(5.3312, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "284000\n",
      "28700 tensor(5.4663, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "285000\n",
      "28800 tensor(5.5155, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "286000\n",
      "28900 tensor(5.3872, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "287000\n",
      "29000 tensor(5.3017, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "288000\n",
      "29100 tensor(5.0583, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "289000\n",
      "29200 tensor(5.2099, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "290000\n",
      "29300 tensor(5.4934, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "291000\n",
      "29400 tensor(5.6202, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "292000\n",
      "29500 tensor(5.4016, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "293000\n",
      "29600 tensor(5.2601, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "294000\n",
      "29700 tensor(5.4038, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "295000\n",
      "29800 tensor(5.2475, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "296000\n",
      "29900 tensor(5.4960, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "297000\n",
      "30000 tensor(5.2438, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "298000\n",
      "30100 tensor(5.3221, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "299000\n",
      "30200 tensor(5.2686, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "300000\n",
      "30300 tensor(5.3735, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "301000\n",
      "30400 tensor(5.2057, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "302000\n",
      "30500 tensor(5.3767, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "30600 tensor(5.3515, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "303000\n",
      "30700 tensor(5.3841, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "304000\n",
      "30800 tensor(5.3889, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "305000\n",
      "30900 tensor(5.4117, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "306000\n",
      "31000 tensor(5.5205, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "307000\n",
      "31100 tensor(5.1742, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "308000\n",
      "31200 tensor(5.2173, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "309000\n",
      "31300 tensor(5.4785, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "310000\n",
      "31400 tensor(5.2577, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "311000\n",
      "31500 tensor(5.4429, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "312000\n",
      "31600 tensor(5.4289, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "313000\n",
      "31700 tensor(5.3961, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "314000\n",
      "31800 tensor(5.4999, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "315000\n",
      "31900 tensor(5.1248, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "316000\n",
      "32000 tensor(5.3122, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "317000\n",
      "32100 tensor(5.1931, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "318000\n",
      "32200 tensor(5.5096, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "319000\n",
      "32300 tensor(5.4973, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "320000\n",
      "32400 tensor(5.4742, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "321000\n",
      "32500 tensor(5.2964, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "322000\n",
      "32600 tensor(5.4063, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "323000\n",
      "32700 tensor(5.3369, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "324000\n",
      "32800 tensor(5.5636, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "325000\n",
      "32900 tensor(5.4245, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "326000\n",
      "33000 tensor(5.2032, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "327000\n",
      "33100 tensor(5.4095, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "328000\n",
      "33200 tensor(5.5071, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "329000\n",
      "33300 tensor(5.2729, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "330000\n",
      "33400 tensor(5.5492, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "331000\n",
      "33500 tensor(5.3701, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "332000\n",
      "33600 tensor(5.3223, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "333000\n",
      "33700 tensor(5.3725, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "334000\n",
      "33800 tensor(5.4572, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "335000\n",
      "33900 tensor(5.1889, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "336000\n",
      "34000 tensor(5.4090, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "337000\n",
      "34100 tensor(5.3798, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "338000\n",
      "34200 tensor(5.4259, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "339000\n",
      "34300 tensor(5.2132, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "340000\n",
      "34400 tensor(5.6692, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "341000\n",
      "34500 tensor(5.5324, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "342000\n",
      "34600 tensor(5.4271, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "343000\n",
      "34700 tensor(5.4978, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "344000\n",
      "34800 tensor(5.5230, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "345000\n",
      "34900 tensor(5.5652, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "346000\n",
      "35000 tensor(5.5478, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "347000\n",
      "35100 tensor(5.3700, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "348000\n",
      "35200 tensor(5.2958, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "349000\n",
      "35300 tensor(5.5219, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "350000\n",
      "35400 tensor(5.1702, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "351000\n",
      "35500 tensor(5.2604, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "352000\n",
      "35600 tensor(5.3821, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "353000\n",
      "35700 tensor(5.2551, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "354000\n",
      "35800 tensor(5.3840, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "355000\n",
      "35900 tensor(5.3635, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "356000\n",
      "36000 tensor(5.1400, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "357000\n",
      "36100 tensor(5.5134, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "358000\n",
      "36200 tensor(5.3632, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "359000\n",
      "36300 tensor(5.6461, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "360000\n",
      "36400 tensor(5.3415, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "361000\n",
      "36500 tensor(5.3659, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "362000\n",
      "36600 tensor(5.3874, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "363000\n",
      "36700 tensor(5.1886, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "364000\n",
      "36800 tensor(5.2958, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "365000\n",
      "36900 tensor(5.4094, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "366000\n",
      "37000 tensor(5.3023, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "367000\n",
      "37100 tensor(5.3287, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "368000\n",
      "37200 tensor(5.3996, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "369000\n",
      "37300 tensor(5.3001, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "370000\n",
      "37400 tensor(5.6516, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "371000\n",
      "37500 tensor(5.3366, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "372000\n",
      "37600 tensor(5.3282, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "373000\n",
      "37700 tensor(5.5061, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "374000\n",
      "37800 tensor(5.3408, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "375000\n",
      "37900 tensor(5.3203, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "376000\n",
      "38000 tensor(5.3996, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "377000\n",
      "38100 tensor(5.4133, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "378000\n",
      "38200 tensor(5.4262, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "379000\n",
      "38300 tensor(5.3305, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "380000\n",
      "38400 tensor(5.3983, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "381000\n",
      "38500 tensor(5.4246, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "382000\n",
      "38600 tensor(5.3713, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "383000\n",
      "38700 tensor(5.3634, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "384000\n",
      "38800 tensor(5.4504, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "385000\n",
      "38900 tensor(5.5273, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "386000\n",
      "39000 tensor(5.2229, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "387000\n",
      "39100 tensor(5.4503, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "388000\n",
      "39200 tensor(5.5406, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "389000\n",
      "39300 tensor(5.3640, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "390000\n",
      "39400 tensor(5.4311, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "391000\n",
      "39500 tensor(5.5292, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "392000\n",
      "39600 tensor(5.2217, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "393000\n",
      "39700 tensor(5.2121, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "394000\n",
      "39800 tensor(5.3415, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "395000\n",
      "39900 tensor(5.1605, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "396000\n",
      "40000 tensor(5.2472, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "397000\n",
      "40100 tensor(5.3351, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "398000\n",
      "40200 tensor(5.3198, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "399000\n",
      "40300 tensor(5.3862, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "40400 tensor(5.3946, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "400000\n",
      "40500 tensor(5.3120, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "401000\n",
      "40600 tensor(5.3741, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "402000\n",
      "40700 tensor(5.4199, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "403000\n",
      "40800 tensor(5.3702, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "404000\n",
      "40900 tensor(5.3212, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "405000\n",
      "41000 tensor(5.3683, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "406000\n",
      "41100 tensor(5.3491, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "407000\n",
      "41200 tensor(5.2400, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "408000\n",
      "41300 tensor(5.3728, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "409000\n",
      "41400 tensor(5.2643, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "410000\n",
      "41500 tensor(5.4064, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "411000\n",
      "41600 tensor(5.3238, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "412000\n",
      "41700 tensor(5.3469, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "413000\n",
      "41800 tensor(5.3432, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "414000\n",
      "41900 tensor(5.3521, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "415000\n",
      "42000 tensor(5.5087, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "416000\n",
      "42100 tensor(5.2556, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "417000\n",
      "42200 tensor(5.3407, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "418000\n",
      "42300 tensor(5.4058, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "419000\n",
      "42400 tensor(5.2231, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "420000\n",
      "42500 tensor(5.3912, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "421000\n",
      "42600 tensor(5.1878, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "422000\n",
      "42700 tensor(5.4955, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "423000\n",
      "42800 tensor(5.4193, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "424000\n",
      "42900 tensor(5.2662, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "425000\n",
      "43000 tensor(5.4093, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "426000\n",
      "43100 tensor(5.4089, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "427000\n",
      "43200 tensor(5.2223, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "428000\n",
      "43300 tensor(5.2456, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "429000\n",
      "43400 tensor(5.4129, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "430000\n",
      "43500 tensor(5.1283, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "431000\n",
      "43600 tensor(5.3275, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "432000\n",
      "epoch: = 4\n",
      "0 tensor(5.3172, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "1000\n",
      "100 tensor(5.3864, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "200 tensor(5.2618, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "2000\n",
      "300 tensor(5.2652, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "3000\n",
      "400 tensor(5.2749, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "4000\n",
      "500 tensor(5.4347, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "5000\n",
      "600 tensor(5.2271, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "6000\n",
      "700 tensor(5.5396, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "7000\n",
      "800 tensor(5.1379, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "8000\n",
      "900 tensor(5.3861, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "9000\n",
      "1000 tensor(5.2629, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "10000\n",
      "1100 tensor(5.4575, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "11000\n",
      "1200 tensor(5.4936, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "12000\n",
      "1300 tensor(5.4281, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "13000\n",
      "1400 tensor(5.4186, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "14000\n",
      "1500 tensor(5.5070, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "15000\n",
      "1600 tensor(5.1769, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "16000\n",
      "1700 tensor(5.2856, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "17000\n",
      "1800 tensor(5.2827, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "18000\n",
      "1900 tensor(5.2544, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "19000\n",
      "2000 tensor(5.3218, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "20000\n",
      "2100 tensor(5.4549, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "21000\n",
      "2200 tensor(5.2864, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "22000\n",
      "2300 tensor(5.3145, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "23000\n",
      "2400 tensor(5.2987, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "24000\n",
      "2500 tensor(5.3498, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "25000\n",
      "2600 tensor(5.3730, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "26000\n",
      "2700 tensor(5.4017, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "27000\n",
      "2800 tensor(5.4255, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "28000\n",
      "2900 tensor(5.5475, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "29000\n",
      "3000 tensor(5.2988, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "30000\n",
      "3100 tensor(5.3753, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "31000\n",
      "3200 tensor(5.4049, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "32000\n",
      "3300 tensor(5.3206, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "33000\n",
      "3400 tensor(5.2159, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "34000\n",
      "3500 tensor(5.3423, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "35000\n",
      "3600 tensor(5.3717, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "36000\n",
      "3700 tensor(5.3042, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "37000\n",
      "3800 tensor(5.3258, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "38000\n",
      "3900 tensor(5.1989, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "39000\n",
      "4000 tensor(5.2650, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "40000\n",
      "4100 tensor(5.3953, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "41000\n",
      "4200 tensor(5.4542, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "42000\n",
      "4300 tensor(5.3466, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "43000\n",
      "4400 tensor(5.4222, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "44000\n",
      "4500 tensor(5.2254, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "45000\n",
      "4600 tensor(5.5610, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "46000\n",
      "4700 tensor(5.2753, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "47000\n",
      "4800 tensor(5.4028, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "48000\n",
      "4900 tensor(5.4516, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "49000\n",
      "5000 tensor(5.4464, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "50000\n",
      "5100 tensor(5.5018, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "51000\n",
      "5200 tensor(5.5194, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "52000\n",
      "5300 tensor(5.5077, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "53000\n",
      "5400 tensor(5.3746, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "54000\n",
      "5500 tensor(5.4847, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "55000\n",
      "5600 tensor(5.2664, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "56000\n",
      "5700 tensor(5.5265, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "57000\n",
      "5800 tensor(5.5101, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "58000\n",
      "5900 tensor(5.4513, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "59000\n",
      "6000 tensor(5.3554, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "60000\n",
      "6100 tensor(5.3616, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "61000\n",
      "6200 tensor(5.5360, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "62000\n",
      "6300 tensor(5.3952, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "63000\n",
      "6400 tensor(5.3132, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "64000\n",
      "6500 tensor(5.1732, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "65000\n",
      "6600 tensor(5.3505, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "66000\n",
      "6700 tensor(5.2919, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "67000\n",
      "6800 tensor(5.5064, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "68000\n",
      "6900 tensor(5.4881, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "69000\n",
      "7000 tensor(5.3978, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "70000\n",
      "7100 tensor(5.2030, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "71000\n",
      "7200 tensor(5.2738, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "72000\n",
      "7300 tensor(5.5317, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "73000\n",
      "7400 tensor(5.3487, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "74000\n",
      "7500 tensor(5.2133, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "75000\n",
      "7600 tensor(5.2878, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "76000\n",
      "7700 tensor(5.3644, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "77000\n",
      "7800 tensor(5.4711, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "78000\n",
      "7900 tensor(5.1445, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "79000\n",
      "8000 tensor(5.2138, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "80000\n",
      "8100 tensor(5.3741, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "81000\n",
      "8200 tensor(5.3893, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "82000\n",
      "8300 tensor(5.3492, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "83000\n",
      "8400 tensor(5.4797, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "84000\n",
      "8500 tensor(5.4501, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "85000\n",
      "8600 tensor(5.4600, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "86000\n",
      "8700 tensor(5.5758, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "87000\n",
      "8800 tensor(5.4493, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "88000\n",
      "8900 tensor(5.3035, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "89000\n",
      "9000 tensor(5.4164, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "90000\n",
      "9100 tensor(5.4273, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "91000\n",
      "9200 tensor(5.3343, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "92000\n",
      "9300 tensor(5.3797, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "93000\n",
      "9400 tensor(5.2260, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "94000\n",
      "9500 tensor(5.3006, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "95000\n",
      "9600 tensor(5.4211, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "96000\n",
      "9700 tensor(5.3615, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "97000\n",
      "9800 tensor(5.4089, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "9900 tensor(5.2200, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "98000\n",
      "10000 tensor(5.4428, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "99000\n",
      "10100 tensor(5.3219, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "100000\n",
      "10200 tensor(5.2692, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "101000\n",
      "10300 tensor(5.3854, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "102000\n",
      "10400 tensor(5.3984, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "103000\n",
      "10500 tensor(5.4516, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "104000\n",
      "10600 tensor(5.5380, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "105000\n",
      "10700 tensor(5.3724, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "106000\n",
      "10800 tensor(5.3862, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "107000\n",
      "10900 tensor(5.3102, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "108000\n",
      "11000 tensor(5.5487, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "109000\n",
      "11100 tensor(5.1684, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "110000\n",
      "11200 tensor(5.3303, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "111000\n",
      "11300 tensor(5.3537, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "112000\n",
      "11400 tensor(5.3064, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "113000\n",
      "11500 tensor(5.3775, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "114000\n",
      "11600 tensor(5.3649, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "115000\n",
      "11700 tensor(5.2920, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "116000\n",
      "11800 tensor(5.4908, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "117000\n",
      "11900 tensor(5.3293, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "118000\n",
      "12000 tensor(5.3926, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "119000\n",
      "12100 tensor(5.4045, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "120000\n",
      "12200 tensor(5.4246, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "121000\n",
      "12300 tensor(5.5096, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "122000\n",
      "12400 tensor(5.3884, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "123000\n",
      "12500 tensor(5.3057, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "124000\n",
      "12600 tensor(5.3466, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "125000\n",
      "12700 tensor(5.2898, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "126000\n",
      "12800 tensor(5.4714, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "127000\n",
      "12900 tensor(5.3255, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "128000\n",
      "13000 tensor(5.1438, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "129000\n",
      "13100 tensor(5.3498, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "130000\n",
      "13200 tensor(5.3890, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "131000\n",
      "13300 tensor(5.2710, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "132000\n",
      "13400 tensor(5.3541, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "133000\n",
      "13500 tensor(5.3156, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "134000\n",
      "13600 tensor(5.3957, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "135000\n",
      "13700 tensor(5.3548, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "136000\n",
      "13800 tensor(5.3577, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "137000\n",
      "13900 tensor(5.2122, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "138000\n",
      "14000 tensor(5.4587, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "139000\n",
      "14100 tensor(5.1704, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "140000\n",
      "14200 tensor(5.4419, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "141000\n",
      "14300 tensor(5.4142, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "142000\n",
      "14400 tensor(5.3058, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "143000\n",
      "14500 tensor(5.4082, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "144000\n",
      "14600 tensor(5.4414, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "145000\n",
      "14700 tensor(5.3177, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "146000\n",
      "14800 tensor(5.7665, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "147000\n",
      "14900 tensor(5.4171, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "148000\n",
      "15000 tensor(5.2698, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "149000\n",
      "15100 tensor(5.4915, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "150000\n",
      "15200 tensor(5.3576, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "151000\n",
      "15300 tensor(5.3567, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "152000\n",
      "15400 tensor(5.2379, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "153000\n",
      "15500 tensor(5.4092, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "154000\n",
      "15600 tensor(5.4042, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "155000\n",
      "15700 tensor(5.3017, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "156000\n",
      "15800 tensor(5.2188, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "157000\n",
      "15900 tensor(5.3497, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "158000\n",
      "16000 tensor(5.1718, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "159000\n",
      "16100 tensor(5.2145, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "160000\n",
      "16200 tensor(5.5591, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "161000\n",
      "16300 tensor(5.3864, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "162000\n",
      "16400 tensor(5.3719, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "163000\n",
      "16500 tensor(5.4842, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "164000\n",
      "16600 tensor(5.3329, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "165000\n",
      "16700 tensor(5.3130, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "166000\n",
      "16800 tensor(5.3903, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "167000\n",
      "16900 tensor(5.1551, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "168000\n",
      "17000 tensor(5.4229, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "169000\n",
      "17100 tensor(5.2686, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "170000\n",
      "17200 tensor(5.4099, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "171000\n",
      "17300 tensor(5.4198, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "172000\n",
      "17400 tensor(5.2162, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "173000\n",
      "17500 tensor(5.2385, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "174000\n",
      "17600 tensor(5.2786, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "175000\n",
      "17700 tensor(5.2576, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "176000\n",
      "17800 tensor(5.3158, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "177000\n",
      "17900 tensor(5.2105, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "178000\n",
      "18000 tensor(5.4627, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "179000\n",
      "18100 tensor(5.3966, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "180000\n",
      "18200 tensor(5.3108, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "181000\n",
      "18300 tensor(5.3148, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "182000\n",
      "18400 tensor(5.3321, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "183000\n",
      "18500 tensor(5.3171, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "184000\n",
      "18600 tensor(5.5247, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "185000\n",
      "18700 tensor(5.3469, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "186000\n",
      "18800 tensor(5.5092, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "187000\n",
      "18900 tensor(5.3902, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "188000\n",
      "19000 tensor(5.3904, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "189000\n",
      "19100 tensor(5.5019, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "190000\n",
      "19200 tensor(5.3838, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "191000\n",
      "19300 tensor(5.4674, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "192000\n",
      "19400 tensor(5.3223, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "193000\n",
      "19500 tensor(5.3235, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "194000\n",
      "19600 tensor(5.3589, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "195000\n",
      "19700 tensor(5.4063, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "196000\n",
      "19800 tensor(5.2838, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "197000\n",
      "19900 tensor(5.2807, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "198000\n",
      "20000 tensor(5.3038, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "199000\n",
      "20100 tensor(5.2397, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "200000\n",
      "20200 tensor(5.1723, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "201000\n",
      "20300 tensor(5.2827, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "20400 tensor(5.3245, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "202000\n",
      "20500 tensor(5.3303, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "203000\n",
      "20600 tensor(5.5211, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "204000\n",
      "20700 tensor(5.3629, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "205000\n",
      "20800 tensor(5.1882, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "206000\n",
      "20900 tensor(5.4671, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "207000\n",
      "21000 tensor(5.3110, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "208000\n",
      "21100 tensor(5.3181, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "209000\n",
      "21200 tensor(5.1968, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "210000\n",
      "21300 tensor(5.3940, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "211000\n",
      "21400 tensor(5.4308, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "212000\n",
      "21500 tensor(5.2127, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "213000\n",
      "21600 tensor(5.2003, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "214000\n",
      "21700 tensor(5.1881, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "215000\n",
      "21800 tensor(5.3180, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "216000\n",
      "21900 tensor(5.3197, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "217000\n",
      "22000 tensor(5.3005, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "218000\n",
      "22100 tensor(5.1776, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "219000\n",
      "22200 tensor(5.0509, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "220000\n",
      "22300 tensor(5.4807, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "221000\n",
      "22400 tensor(5.2040, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "222000\n",
      "22500 tensor(5.2161, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "223000\n",
      "22600 tensor(5.4083, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "224000\n",
      "22700 tensor(5.1619, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "225000\n",
      "22800 tensor(5.4301, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "226000\n",
      "22900 tensor(5.4791, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "227000\n",
      "23000 tensor(5.3785, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "228000\n",
      "23100 tensor(5.3705, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "229000\n",
      "23200 tensor(5.3633, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "230000\n",
      "23300 tensor(5.4443, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "231000\n",
      "23400 tensor(5.4496, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "232000\n",
      "23500 tensor(5.2961, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "233000\n",
      "23600 tensor(5.2603, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "234000\n",
      "23700 tensor(5.2793, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "235000\n",
      "23800 tensor(5.1461, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "236000\n",
      "23900 tensor(5.2376, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "237000\n",
      "24000 tensor(5.2269, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "238000\n",
      "24100 tensor(5.3154, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "239000\n",
      "24200 tensor(5.4852, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "240000\n",
      "24300 tensor(5.3785, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "241000\n",
      "24400 tensor(5.5053, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "242000\n",
      "24500 tensor(5.2987, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "243000\n",
      "24600 tensor(5.4275, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "244000\n",
      "24700 tensor(5.3283, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "245000\n",
      "24800 tensor(5.3707, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "246000\n",
      "24900 tensor(5.4294, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "247000\n",
      "25000 tensor(5.4479, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "248000\n",
      "25100 tensor(5.3629, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "249000\n",
      "25200 tensor(5.3849, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "250000\n",
      "25300 tensor(5.4124, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "251000\n",
      "25400 tensor(5.3932, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "252000\n",
      "25500 tensor(5.2893, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "253000\n",
      "25600 tensor(5.5512, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "254000\n",
      "25700 tensor(5.3227, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "255000\n",
      "25800 tensor(5.4217, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "256000\n",
      "25900 tensor(5.3637, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "257000\n",
      "26000 tensor(5.3632, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "258000\n",
      "26100 tensor(5.2841, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "259000\n",
      "26200 tensor(5.2107, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "260000\n",
      "26300 tensor(5.4024, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "261000\n",
      "26400 tensor(5.4410, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "262000\n",
      "26500 tensor(5.1685, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "263000\n",
      "26600 tensor(5.5023, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "264000\n",
      "26700 tensor(5.3654, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "265000\n",
      "26800 tensor(5.5407, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "266000\n",
      "26900 tensor(5.3000, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "267000\n",
      "27000 tensor(5.2141, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "268000\n",
      "27100 tensor(5.2490, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "269000\n",
      "27200 tensor(5.2850, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "270000\n",
      "27300 tensor(5.4811, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "271000\n",
      "27400 tensor(5.3561, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "272000\n",
      "27500 tensor(5.2602, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "273000\n",
      "27600 tensor(5.5429, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "274000\n",
      "27700 tensor(5.3794, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "275000\n",
      "27800 tensor(5.3792, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "276000\n",
      "27900 tensor(5.4873, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "277000\n",
      "28000 tensor(5.3454, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "278000\n",
      "28100 tensor(5.3113, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "279000\n",
      "28200 tensor(5.4785, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "280000\n",
      "28300 tensor(5.1013, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "281000\n",
      "28400 tensor(5.5403, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "282000\n",
      "28500 tensor(5.3676, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "283000\n",
      "28600 tensor(5.3108, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "284000\n",
      "28700 tensor(5.4403, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "285000\n",
      "28800 tensor(5.4926, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "286000\n",
      "28900 tensor(5.3638, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "287000\n",
      "29000 tensor(5.2819, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "288000\n",
      "29100 tensor(5.0362, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "289000\n",
      "29200 tensor(5.1871, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "290000\n",
      "29300 tensor(5.4697, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "291000\n",
      "29400 tensor(5.5909, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "292000\n",
      "29500 tensor(5.3807, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "293000\n",
      "29600 tensor(5.2398, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "294000\n",
      "29700 tensor(5.3690, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "295000\n",
      "29800 tensor(5.2220, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "296000\n",
      "29900 tensor(5.4597, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "297000\n",
      "30000 tensor(5.2205, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "298000\n",
      "30100 tensor(5.3061, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "299000\n",
      "30200 tensor(5.2432, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "300000\n",
      "30300 tensor(5.3527, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "301000\n",
      "30400 tensor(5.1823, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "302000\n",
      "30500 tensor(5.3526, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "30600 tensor(5.3318, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "303000\n",
      "30700 tensor(5.3634, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "304000\n",
      "30800 tensor(5.3571, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "305000\n",
      "30900 tensor(5.3875, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "306000\n",
      "31000 tensor(5.4983, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "307000\n",
      "31100 tensor(5.1554, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "308000\n",
      "31200 tensor(5.1952, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "309000\n",
      "31300 tensor(5.4546, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "310000\n",
      "31400 tensor(5.2307, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "311000\n",
      "31500 tensor(5.4188, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "312000\n",
      "31600 tensor(5.4085, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "313000\n",
      "31700 tensor(5.3744, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "314000\n",
      "31800 tensor(5.4766, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "315000\n",
      "31900 tensor(5.1062, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "316000\n",
      "32000 tensor(5.2924, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "317000\n",
      "32100 tensor(5.1728, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "318000\n",
      "32200 tensor(5.4863, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "319000\n",
      "32300 tensor(5.4748, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "320000\n",
      "32400 tensor(5.4518, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "321000\n",
      "32500 tensor(5.2752, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "322000\n",
      "32600 tensor(5.3822, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "323000\n",
      "32700 tensor(5.3088, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "324000\n",
      "32800 tensor(5.5403, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "325000\n",
      "32900 tensor(5.4000, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "326000\n",
      "33000 tensor(5.1837, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "327000\n",
      "33100 tensor(5.3888, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "328000\n",
      "33200 tensor(5.4849, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "329000\n",
      "33300 tensor(5.2471, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "330000\n",
      "33400 tensor(5.5246, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "331000\n",
      "33500 tensor(5.3479, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "332000\n",
      "33600 tensor(5.3043, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "333000\n",
      "33700 tensor(5.3487, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "334000\n",
      "33800 tensor(5.4368, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "335000\n",
      "33900 tensor(5.1620, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "336000\n",
      "34000 tensor(5.3873, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "337000\n",
      "34100 tensor(5.3545, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "338000\n",
      "34200 tensor(5.4001, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "339000\n",
      "34300 tensor(5.1902, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "340000\n",
      "34400 tensor(5.6453, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "341000\n",
      "34500 tensor(5.5124, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "342000\n",
      "34600 tensor(5.4069, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "343000\n",
      "34700 tensor(5.4734, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "344000\n",
      "34800 tensor(5.5014, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "345000\n",
      "34900 tensor(5.5412, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "346000\n",
      "35000 tensor(5.5132, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "347000\n",
      "35100 tensor(5.3455, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "348000\n",
      "35200 tensor(5.2694, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "349000\n",
      "35300 tensor(5.4988, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "350000\n",
      "35400 tensor(5.1485, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "351000\n",
      "35500 tensor(5.2299, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "352000\n",
      "35600 tensor(5.3643, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "353000\n",
      "35700 tensor(5.2247, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "354000\n",
      "35800 tensor(5.3615, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "355000\n",
      "35900 tensor(5.3453, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "356000\n",
      "36000 tensor(5.1217, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "357000\n",
      "36100 tensor(5.4909, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "358000\n",
      "36200 tensor(5.3382, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "359000\n",
      "36300 tensor(5.6225, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "360000\n",
      "36400 tensor(5.3167, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "361000\n",
      "36500 tensor(5.3458, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "362000\n",
      "36600 tensor(5.3608, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "363000\n",
      "36700 tensor(5.1660, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "364000\n",
      "36800 tensor(5.2737, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "365000\n",
      "36900 tensor(5.3883, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "366000\n",
      "37000 tensor(5.2783, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "367000\n",
      "37100 tensor(5.3110, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "368000\n",
      "37200 tensor(5.3794, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "369000\n",
      "37300 tensor(5.2802, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "370000\n",
      "37400 tensor(5.6133, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "371000\n",
      "37500 tensor(5.3138, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "372000\n",
      "37600 tensor(5.3083, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "373000\n",
      "37700 tensor(5.4860, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "374000\n",
      "37800 tensor(5.3216, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "375000\n",
      "37900 tensor(5.2969, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "376000\n",
      "38000 tensor(5.3759, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "377000\n",
      "38100 tensor(5.3914, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "378000\n",
      "38200 tensor(5.4089, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "379000\n",
      "38300 tensor(5.3068, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "380000\n",
      "38400 tensor(5.3798, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "381000\n",
      "38500 tensor(5.4051, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "382000\n",
      "38600 tensor(5.3471, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "383000\n",
      "38700 tensor(5.3415, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "384000\n",
      "38800 tensor(5.4310, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "385000\n",
      "38900 tensor(5.5029, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "386000\n",
      "39000 tensor(5.2021, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "387000\n",
      "39100 tensor(5.4283, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "388000\n",
      "39200 tensor(5.5158, device='cuda:0', grad_fn=<NllLossBackward0>)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "389000\n",
      "39300 tensor(5.3452, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "390000\n",
      "39400 tensor(5.4111, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "391000\n",
      "39500 tensor(5.4969, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "392000\n",
      "39600 tensor(5.1952, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "393000\n",
      "39700 tensor(5.1946, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "394000\n",
      "39800 tensor(5.3234, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "395000\n",
      "39900 tensor(5.1354, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "396000\n",
      "40000 tensor(5.2210, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "397000\n",
      "40100 tensor(5.3133, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "398000\n",
      "40200 tensor(5.2990, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "399000\n",
      "40300 tensor(5.3684, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "40400 tensor(5.3700, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "400000\n",
      "40500 tensor(5.2911, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "401000\n",
      "40600 tensor(5.3497, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "402000\n",
      "40700 tensor(5.3981, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "403000\n",
      "40800 tensor(5.3436, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "404000\n",
      "40900 tensor(5.2978, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "405000\n",
      "41000 tensor(5.3420, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "406000\n",
      "41100 tensor(5.3342, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "407000\n",
      "41200 tensor(5.2226, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "408000\n",
      "41300 tensor(5.3573, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "409000\n",
      "41400 tensor(5.2448, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "410000\n",
      "41500 tensor(5.3863, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "411000\n",
      "41600 tensor(5.3051, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "412000\n",
      "41700 tensor(5.3294, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "413000\n",
      "41800 tensor(5.3191, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "414000\n",
      "41900 tensor(5.3289, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "415000\n",
      "42000 tensor(5.4860, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "416000\n",
      "42100 tensor(5.2358, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "417000\n",
      "42200 tensor(5.3253, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "418000\n",
      "42300 tensor(5.3869, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "419000\n",
      "42400 tensor(5.2062, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "420000\n",
      "42500 tensor(5.3712, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "421000\n",
      "42600 tensor(5.1718, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "422000\n",
      "42700 tensor(5.4735, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "423000\n",
      "42800 tensor(5.3973, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "424000\n",
      "42900 tensor(5.2447, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "425000\n",
      "43000 tensor(5.3896, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "426000\n",
      "43100 tensor(5.3916, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "427000\n",
      "43200 tensor(5.2044, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "428000\n",
      "43300 tensor(5.2167, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "429000\n",
      "43400 tensor(5.3933, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "430000\n",
      "43500 tensor(5.1078, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "431000\n",
      "43600 tensor(5.3045, device='cuda:0', grad_fn=<NllLossBackward0>)\n",
      "432000\n"
     ]
    }
   ],
   "source": [
    "data = DataLoader(train_dataset, batch_size=batch_s)\n",
    "optimizer = torch.optim.Adam(model.parameters(), lr=learning_rate)\n",
    "criterion = torch.nn.NLLLoss()\n",
    "torch.cuda.empty_cache()\n",
    "gc.collect()\n",
    "\n",
    "model.load_state_dict(torch.load('model-bigram_final.bin'))\n",
    "for i in range(1, epochs+1):\n",
    "    print('epoch: =', i)\n",
    "    model.train()\n",
    "    step = 0\n",
    "    for x, y in data: # prev, predicting, following words\n",
    "       x = x.to(device)\n",
    "       y = y.to(device)\n",
    "       optimizer.zero_grad()\n",
    "       ypredicted = model(x) #previous, following word\n",
    "       loss = criterion(torch.log(ypredicted), y)\n",
    "       if step % 100 == 0:\n",
    "          print(step, loss)\n",
    "       step += 1\n",
    "       loss.backward()\n",
    "       optimizer.step()\n",
    "    torch.save(model.state_dict(), f'model-bigram_2nd-run{i}.bin')    \n",
    "torch.save(model.state_dict(), f'model-bigram_final.bin')    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('be', 11, 0.2570849657058716),\n",
       " ('<unk>', 0, 0.07411641627550125),\n",
       " ('not', 22, 0.05940083786845207),\n",
       " ('have', 28, 0.02751326560974121),\n",
       " ('bo', 167, 0.014936885796487331),\n",
       " ('make', 116, 0.013943656347692013),\n",
       " ('give', 193, 0.011286991648375988),\n",
       " ('take', 153, 0.011171611957252026),\n",
       " ('do', 86, 0.010088067501783371),\n",
       " ('he', 20, 0.009703895077109337)]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device = 'cuda'\n",
    "torch.cuda.empty_cache()\n",
    "model = SimpleBigramNeuralLanguageModel(vocab_size, embed_size).to(device)\n",
    "model.load_state_dict(torch.load(f'model-bigram_final.bin'))\n",
    "model.eval()\n",
    "\n",
    "ixs = torch.tensor(vocab.forward(['will'])).to(device)\n",
    "\n",
    "out = model(ixs)\n",
    "top = torch.topk(out[0], 10)\n",
    "top_indices = top.indices.tolist()\n",
    "top_probs = top.values.tolist()\n",
    "top_words = vocab.lookup_tokens(top_indices)\n",
    "list(zip(top_words, top_indices, top_probs))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('<unk>', 0, 0.19996878504753113),\n",
       " ('and', 3, 0.05288130044937134),\n",
       " ('of', 2, 0.042051784694194794),\n",
       " ('the', 1, 0.026572922244668007),\n",
       " ('to', 4, 0.022689413279294968),\n",
       " ('in', 6, 0.015904497355222702),\n",
       " ('The', 17, 0.012827681377530098),\n",
       " ('a', 5, 0.00961760152131319),\n",
       " ('for', 8, 0.008938422426581383),\n",
       " ('</s>', 32, 0.00840282253921032)]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vocab = train_dataset.vocab\n",
    "ixs = torch.tensor(vocab.forward(['cerned.'])).to(device)\n",
    "\n",
    "out = model(ixs)\n",
    "top = torch.topk(out[0], 10)\n",
    "top_indices = top.indices.tolist()\n",
    "top_probs = top.values.tolist()\n",
    "top_words = vocab.lookup_tokens(top_indices)\n",
    "list(zip(top_words, top_indices, top_probs))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('<unk>', 0, 1.0),\n",
       " ('particular,', 14538, 0.24527804553508759),\n",
       " ('revolution.', 20446, 0.23776617646217346),\n",
       " ('Territory.', 14189, 0.23417341709136963),\n",
       " ('or-', 2261, 0.22888363897800446),\n",
       " ('3', 479, 0.2288265973329544),\n",
       " ('speak.', 13722, 0.2252315878868103),\n",
       " ('attend.', 19397, 0.22110989689826965),\n",
       " ('say,', 1455, 0.22106117010116577),\n",
       " ('Lee.', 15326, 0.21764159202575684)]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cos = nn.CosineSimilarity(dim=1, eps=1e-6)\n",
    "\n",
    "embeddings = model.model[0].weight\n",
    "\n",
    "vec = embeddings[vocab['cerned.']]\n",
    "\n",
    "similarities = cos(vec, embeddings)\n",
    "\n",
    "top = torch.topk(similarities, 10)\n",
    "\n",
    "top_indices = top.indices.tolist()\n",
    "top_probs = top.values.tolist()\n",
    "top_words = vocab.lookup_tokens(top_indices)\n",
    "list(zip(top_words, top_indices, top_probs))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "\n",
    "vocab = train_dataset.vocab\n",
    "# ixs = torch.tensor(vocab.forward(['a'])).to(device)\n",
    "ixs = torch.tensor(vocab.forward(['of'])).to(device)\n",
    "# ixs = torch.tensor(vocab.forward(['that'])).to(device)\n",
    "# ixs = torch.tensor(vocab.forward(['church'])).to(device)\n",
    "# ixs = torch.tensor(vocab.forward(['wait'])).to(device)\n",
    "\n",
    "out = model(ixs)\n",
    "top = torch.topk(out[0], 10)\n",
    "top_indices = top.indices.tolist()\n",
    "top_probs = top.values.tolist()\n",
    "top_words = vocab.lookup_tokens(top_indices)\n",
    "list(zip(top_words, top_indices, top_probs))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_values_from_model(presc_word, model, vocab, k):\n",
    "    ixs =  torch.tensor(vocab.forward([presc_word])).to(device)\n",
    "    out = model(ixs)\n",
    "    top = torch.topk(out[0], k)\n",
    "    top_indices = top.indices.tolist()\n",
    "    top_probs = top.values.tolist()\n",
    "    top_words = vocab.lookup_tokens(top_indices)\n",
    "    return list(zip(top_words, top_probs))\n",
    "\n",
    "def gonito_format(dic):\n",
    "    tab = summarize_probs_unk(dic)\n",
    "    result = ''\n",
    "    for element in tab[:-1]:\n",
    "        result+=str(element[0])+':'+str(element[1])+'\\t'\n",
    "    result+=':'+ str(tab[-1][1])+'\\n'\n",
    "    return result\n",
    "\n",
    "def summarize_probs_unk(dic):\n",
    "    if '<unk>' in dic.keys():\n",
    "        del dic['<unk>']\n",
    "    probsum = sum(float(val) for key, val in dic.items())\n",
    "#     if \"<unk>\" in dic.keys():\n",
    "#         for key in dic:\n",
    "#             dic[key] = dic[key]/probsum #normalize to 1\n",
    "#         wildcard = dic['<unk>'] \n",
    "#     else: #no unk in first 10 entries\n",
    "    for key in dic:\n",
    "        dic[key] = dic[key]/probsum*(1-wildcard_minweight) ###leave some space for wildcard\n",
    "    wildcard = wildcard_minweight\n",
    "    tab = [(key, val) for key, val in dic.items()]\n",
    "    tab.append(('<unk>', wildcard))\n",
    "    return tab\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Present,\n",
      "maintenance.\n",
      "-\n",
      "on\n",
      "after.\n",
      "included.\n",
      "Delaware.\n",
      "body.\n",
      "route.\n",
      "under\n",
      "bership.\n",
      "call*\n",
      "waterscape.\n",
      "?\n",
      "soles.\n",
      "simple.\"\n",
      "public.\n",
      "them.-\n",
      "months.\n",
      "icy.\"\n",
      "E\n",
      "level.\n",
      "4t—36\n",
      "financial\n",
      "steady.\n",
      "mlnicars.\n",
      "his\n",
      "circumstances.\n",
      "corner.\n",
      "science.\n",
      "insurance.\"\n",
      "fair.\n",
      "scalded.\n",
      "ria.\n",
      "it,\n",
      "mlngtonlan.\n",
      "possession.\n",
      "of\n",
      "pray!”\n",
      "occurred.\n",
      "euro,\n",
      "way.\n",
      "185.\n",
      "destruction.\n",
      "conditions.\n",
      "Hood.\n",
      "s.\n",
      "case.\n",
      "Roy\n",
      "country.\n",
      "ccjulinalf\n",
      "side.\n",
      "against\n",
      "servation.\n",
      "Detroit.\n",
      "North\n",
      "country.\n",
      "\"\n",
      "bones.\n",
      "sold.\n",
      "Brown\n",
      "bill.\n",
      "Springer,\n",
      "gaged.\n",
      "their\n",
      "character.\n",
      "cu\n",
      "former\n",
      "cities.\n",
      "accumulated\n",
      "corporal.\n",
      "^\n",
      "-\n",
      "o\n",
      ".<!.\n",
      "them.\n",
      "wounded.\n",
      "by.\n",
      "of\n",
      "remedy.\n",
      "debate.\n",
      "22-28).\n",
      "\"\n",
      "book.\"\n",
      "contract,\n",
      "presentation.\n",
      "Ellsha\n",
      "praise.\n",
      "opin-\n",
      "proof.\n",
      "lows:\n",
      "come\n",
      "dispensation.\n",
      "point.\n",
      "theirjfnmiliea.\n",
      "ull.\n",
      "Magazine.\n",
      "rnUnil.\n",
      "order.\n",
      "morrow\n",
      "nrL\n",
      "is\n",
      "'death.\n",
      "caucus.\n",
      "east\n",
      "-\n",
      "service.\n",
      "cor-\n",
      "ton\n",
      "answer.\n",
      "bay.\n",
      "same.\n",
      ":\n",
      "Arkansas.\n",
      "lives.\n",
      "the\n",
      "t\n",
      ".1\n",
      "P\n",
      "lems?\n",
      "4,\n",
      "know,'\n",
      "his\n",
      "arouth.\"\n",
      "offer.\n",
      "horizontal!'.\n",
      "two-thir-\n",
      "city.\n",
      "come\n",
      "here.\n",
      "STORY\n",
      "auxiliaries.\n",
      "property.\n",
      "respected.\n",
      "stopped.\n",
      "JanlO\n",
      "honesty.\n",
      "civilian\n",
      "servation.\n",
      "army.\n",
      "Harbaugh.\n",
      "ear.\n",
      "beginning.\n",
      "State%\n",
      "nA.a\n",
      "operation.\n",
      "cash.\n",
      "upon.\n",
      "drugs.\n",
      "kind.\n",
      "out.\n",
      "f\n",
      "st.\n",
      "season.\n",
      "WUK\n",
      "over,\n",
      "offriends.\n",
      "Inger-\n",
      "recovery,-\n",
      "separately.\n",
      "bachelor.\n",
      "offer.\n",
      "Colony.\n",
      "and\n",
      "wi\n",
      "we\n",
      "i\n",
      "them.\n",
      "rons\n",
      "’<>(.\n",
      "descendants\n",
      "notes.\n",
      "Courier.\n",
      "town\n",
      "E\n",
      "awake,\n",
      "d\n",
      "circumstances.\n",
      "board.\n",
      "-\n",
      "union.\n",
      "the\n",
      "en-\n",
      "children\n",
      "Tre--\n",
      "date.\n",
      "Monday.\n",
      "been\n",
      "now\n",
      "slaughter.”\n",
      "art.\n",
      "shape.\n",
      "ai\n",
      "Magazine.\n",
      "License.\n",
      "K\n",
      "P.iU--\n",
      "failures\n",
      "v\n",
      "cause.\n",
      "Russian\n",
      "prayed.\n",
      "tariff.\n",
      "action.\n",
      "direction.\n",
      "Tacomans\n",
      "to-\n",
      "cellence.\n",
      "power\n",
      "25\n",
      "alty.\n",
      "July.\"\n",
      "crime.\n",
      "of\n",
      "conviction.\n",
      "been\n",
      "d\n",
      "unnatural\n",
      "s,\n",
      "-\n",
      "scccdo.\n",
      "e-\n",
      "Harrisburg.\n",
      "cash.\n",
      "Peter\n",
      "OTTirn\n",
      "lor\n",
      "MacAuley.\n",
      "O\n",
      "style.\n",
      "road.\n",
      "YOU.\n",
      "nations.\"\n",
      "r.o\n",
      "made.\n",
      "Democratic.\n",
      "......\n",
      "Ocean.\n",
      "r\n",
      "and\n",
      "bless\n",
      "jaSl\n",
      "Krapt.\n",
      "states.\n",
      "It.\n",
      "river.\n",
      "fag-\n",
      "ShorteU.\n",
      "fabric.\"\n",
      "returned.\n",
      "tra.\n",
      "member.\n",
      "both.\n",
      "I\n",
      "place.\"\n",
      "our\n",
      ".»\n",
      "17\n",
      "1911.\n",
      "an\n",
      "Age.\n",
      "It\n",
      "Gland--\n",
      "tl\n",
      "-\n",
      "be-\n",
      "air.\n",
      "crude.\n",
      "remedy.\n",
      ":\n",
      "Uny\n",
      "tunnel.\n",
      "n\n",
      "Gulf.”\n",
      "war.\n",
      "...\n",
      "contest\n",
      "History.\n",
      "aear.\n",
      "tees.\n",
      "residence.\n",
      "friend,\n",
      "h\n",
      "\"\n",
      "him.\n",
      "government.\n",
      "facilities.\"\n",
      "remedy.\n",
      "r\n",
      "eent\n",
      "law.\n",
      "the\n",
      "Representative\n",
      "w\n",
      "n\n",
      "1S44.\n",
      "adopted.\n",
      "favor.\n",
      "Sciurier.\n",
      "bein\n",
      "market.\n",
      ".1\n",
      "marshes.\n",
      "Newark.\n",
      "years\n",
      "Preeton.\"\n",
      "lti-\n",
      "Trumpet\n",
      "said.\n",
      "appoint\n",
      "oi\n",
      "tbe\n",
      "Federahi,\n",
      "1858.\n",
      "hereof.\n",
      "treatment.\n",
      "dawdling\n",
      "dence.\n",
      "L\n",
      "requirements.\n",
      "in\n",
      "e\n",
      "projects?\n",
      "goes:\n",
      "W\n",
      "w\n",
      "thereof.\n",
      "It\n",
      "injury.\"\n",
      "water.\"\n",
      "played\n",
      "famoti'\n",
      "nnd\n",
      "there.'\n",
      "win.\"\n",
      "used.\n",
      ".by\n",
      "'\n",
      "Woman.\n",
      "viz:\n",
      "causing\n",
      "quarter.\n",
      "i.\n",
      "police.\n",
      "run.\n",
      "children.'\n",
      "debate.\n",
      "Blannerhasset\n",
      "pa-\n",
      "with\n",
      "elbow.\n",
      "J\n",
      "aid.\n",
      "States.\n",
      "graii\n",
      "terday.\n",
      "was\n",
      "mlngo\n",
      "night.\n",
      "reason.\n",
      "Thlinget.\n",
      "111.\n",
      "'\n",
      "anarled.\n",
      "personally.\n",
      "man.\n",
      "nonresidents\n",
      "man.\n",
      "sently.\n",
      "st\n",
      "time.\n",
      "shrine.\n",
      "communicable\n",
      "anf\n",
      "Th.-\n",
      "lose\n",
      "thereon.\n",
      "application.\n",
      "disappear.\n",
      "forfeited.\n",
      "collis-\n",
      "sirl\n",
      "(\n",
      "reconstructed.\n",
      "(ban.)\n",
      "the\n",
      "tion.\n",
      "afte\n",
      "co-urts.\"\n",
      "18W2.\n",
      "attuck.\n",
      "too\n",
      "void.\n",
      "presented,\n",
      "shallton.\n",
      "pnst\n",
      "here.\n",
      "mentioned.\n",
      "session.\n",
      "b\n",
      "s\n",
      "river.\n",
      "capac\n",
      "firm.\n",
      "bombardment,\n",
      "other.\n",
      "worship.\n",
      "The\n",
      "liberties.\n",
      "respousihiliiics,\n",
      "Times.\n",
      "other.\n",
      "following:\n",
      "consid­\n",
      "arroya.\n",
      "owno-\n",
      "is\n",
      "14.\n",
      "name.\n",
      "night\n",
      "adjourned.\n",
      "morning.^\n",
      "paint.\n",
      "post-nat-\n",
      "union.\n",
      "or-\n",
      "Jiuili\n",
      "said.\n",
      "splurge.\n",
      "him.\n",
      "work.\"\n",
      "-te-\n",
      "tfo\n",
      "there\n",
      ".\n",
      "rising.\n",
      "Gen.\n",
      "11\n",
      "yours,\n",
      "Farmer.\n",
      "acquaintanceship,\n",
      "hand.\n",
      "high-cla-\n",
      "nights.\n",
      "AfcC-i-\n",
      "elections.\n",
      "way.\"\n",
      "once.\n",
      ".\n",
      "comfortable.\n",
      "rality.\n",
      ".be-\n",
      "approval,\n",
      "sto-\n",
      "cure.\n",
      "27,000.\n",
      "spoils.”\n",
      "deeds.\n",
      "now\n",
      "1.\n",
      "da\n",
      "...11.11-\n",
      "d\n",
      "berry.\n",
      "congratulations.\n",
      "coast\n",
      "Wednesday\n",
      "ferences.\n",
      "point.\"\n",
      "preservation.\n",
      "gowns.\n",
      "curanhalted.\n",
      "JBarmsT.\n",
      "surroundings.\n",
      "treat\n",
      "ted.\n",
      "tfrith\n",
      "blossoms.\n",
      "here.\"\n",
      "Minnesota.\n",
      "rope.\n",
      "—\n",
      "him:\n",
      "*\n",
      "is”\n",
      "feet\n",
      "music-\n",
      "Newark.\n",
      "given\n",
      "country\n",
      "sick.\n",
      "black.\n",
      "so.\n",
      "archdiocese.\n",
      "form.\"\n",
      "k\n",
      "propose.\n",
      "ed.\n",
      ".\n",
      "healthy.\n",
      "cannons.\n",
      "auxiliaries\n",
      "by\n",
      "to\n",
      "shoute,\n",
      "thel.-\n",
      "\"Short\n",
      "side.\n",
      "Koine.\n",
      "thanks.\n",
      "le.\n",
      "communicable.\n",
      "[Sigued]\n",
      "ion.\n",
      "remedy.\n",
      "Monthly.\n",
      "enterprises.\"\n",
      "nlieht**.i\n",
      "harmed.\"\n",
      "retreat.\n",
      "what\n",
      "d\n",
      "over\n",
      "wounded.\n",
      "-\n",
      ".\n",
      "mobiles.\"\n",
      "mill.\n",
      "college.\n",
      "B.\n",
      "Manassas.\n",
      "fea-\n",
      "fatigue.\n",
      "\"Quartermaster\n",
      "Williams.'\n",
      "visible.\n",
      "Cox,\n",
      "bo\n",
      "stitution.\n",
      "said\n",
      "c-\n",
      "be\n",
      "hat.\n",
      "Sing.\n",
      "penalty.\n",
      "at\n",
      "prison!\n",
      "salt.\n",
      "consider.\"\n",
      "mo\n",
      "TWENTY-EIGHT-\n",
      "Hazen's\n",
      "F\n",
      "Pennsylvania\n",
      "mand.\n",
      "Class.”\n",
      "years.\n",
      "shore.\n",
      "posterity.\n",
      "1000.\n",
      "tern-\n",
      "meeting.\n",
      "logical.\n",
      "said:\n",
      "stamps.\n",
      "trol.\"\n",
      "misdeeds.\n",
      "Middletown.\n",
      "nnd\n",
      "law.\n",
      "sure\n",
      "recover.\n",
      "e\n",
      "move.\n",
      "remedy.\n",
      "days.\n",
      "efficiency.\n",
      "provement.\n",
      "be\n",
      "i\n",
      "fall.\n",
      "East\n",
      "really\n",
      "-\n",
      "uever\n",
      "leg-\n",
      "deny\n",
      "reason\n",
      "ous\n",
      "of\n",
      "agencies.\n",
      "eloped.\"\n",
      "scale.\n",
      "Colonel\n",
      "company.\n",
      "watera\n",
      "|\n",
      "enterprises,\n",
      "to-day.\n",
      "In\n",
      "enuffforme.\n",
      "justly\n",
      "finitum.\"\n",
      "negan.\n",
      "OTTO\n",
      "interests.\n",
      "mo-\n",
      "determined.\n",
      "iasu\n",
      "void.\n",
      "nights\n",
      "come.\"\n",
      ".\n",
      "pild.\n",
      "(\n",
      "her\n",
      "from\n",
      "Russians.\n",
      "things.\n",
      "talk.\n",
      "constituted,\n",
      "lows:\n",
      "dustry.\n",
      "state.\n",
      "sections.\n",
      "cent..\n",
      "cm'\n",
      "mile*.\n",
      "liberty,\n",
      "judgment.\n",
      "a-t\n",
      "illusion.\n",
      "With\n",
      "States.\n",
      "and\n",
      "kind.\n",
      "preference\n",
      "lunch.\n",
      "tary.\n",
      "”\n",
      "the\n",
      "Democratic\n",
      "»ball\n",
      "ribbon.\n",
      "and\n",
      "of\n",
      ".\n",
      "route.\n",
      "prayed.\n",
      "The\n",
      "live.\n",
      "train.\n",
      "holi-\n",
      "follow.\n",
      "a\n",
      "quail.”\n",
      "uuu\n",
      "oath.\n",
      "Cohce.\n",
      "',\n",
      "record.\n",
      "it.'\n",
      "the\n",
      "have\n",
      "Kinley.\n",
      "purchase.\n",
      "violcuce.\n",
      "gates-wor-e\n",
      "placed\n",
      "taste.\n",
      "at-\n",
      "servation.\n",
      "\"\n",
      "views.\n",
      "one.\n",
      "description.\n",
      "lized\n",
      "egress.\n",
      "flames.\n",
      "e\n",
      "704c.\n",
      "height.\n",
      "office.\n",
      "Pa.,\n",
      "preserves\n",
      "voice.\n",
      "Ledger.\n",
      "later.\n",
      "eminent\n",
      "powerfully.\"\n",
      "d\n",
      "g\n",
      "Iremain\n",
      "mi-\n",
      "Constitu-\n",
      "charges.\n",
      "llnllrond\n",
      "llWi\n",
      "n,\n",
      "wel-\n",
      "eternity.\n",
      "of\n",
      "it.\n",
      "on-\n",
      "n-\n",
      "sure.\"\n",
      "n-\n",
      "feform.\n",
      "it.\n",
      "man.\n",
      "popularity.\n",
      "PUGH.\n",
      "dollars.\n",
      "w-elfare.\n",
      "motionless.\n",
      "ago\n",
      "call.\n",
      "emergencies.\n",
      "45.\n",
      "again.\n",
      "to\n",
      "Smith,\n",
      "a\n",
      "h\n",
      "Client\n",
      "for.\n",
      "gone.\n",
      "thereof.\n",
      "per-\n",
      "herlivin?.\n",
      "d\n",
      "Th-\n",
      "them,\n",
      "nbllWiM\n",
      "Measures,\n",
      "dreary,\n",
      "crime.\"\n",
      "^ffsouJ.\n",
      "obser-\n",
      "home.\n",
      "him.\n",
      "bcr.\n",
      "C.\n",
      "were\n",
      "purchaser.\n",
      "the\n",
      "tu-\n",
      "policy.\n",
      "Dalty!\"\n",
      "enl-\n",
      "win.\"\n",
      "dissolution.\n",
      "health.\n",
      "exception.\n",
      "16'h.\n",
      "1859.\n",
      "boat.\n",
      "'\n",
      "home.\n",
      "eausees.\n",
      "Congress.\"\n",
      "few.\n",
      "’'\n",
      "cost:\n",
      "demand.\n",
      "them.\n",
      "Vlr\n",
      "at\n",
      "Saint\n",
      "which\n",
      "all.\n",
      "consider.\n",
      "list.\n",
      "on\n",
      "menu.\n",
      "V.rk.\n",
      "onstrate.\n",
      "nor\n",
      "as\n",
      "county.\"\n",
      "Iowu,\n",
      "grade.\n",
      "organization.\n",
      "Wyman.\n",
      "Unsportsmanlike\n",
      "pioneers.\n",
      "spring.\n",
      "appertaining.\n",
      "porch.\n",
      "James\n",
      "we\n",
      "\"\n",
      "it.\n",
      "hand.\"\n",
      "farew-ell.\"\n",
      "advantages.\n",
      "all\n",
      "-\n",
      "secured\n",
      "attention\n",
      "water\n",
      "igoV\n",
      "State.\n",
      "cattle\n",
      "othor.\n",
      ".more.\n",
      "-\n",
      ".\n",
      ",\n",
      "rest.\n",
      "been,-\n",
      "a»\n",
      "rei\n",
      "cases.\n",
      "is.\n",
      "d\n",
      "aforesaid.\n",
      "determine.\n",
      "him.\n",
      "comic,\n",
      "Inquiry.\n",
      "girls,\n",
      "rations.\n",
      "living.\n",
      "delinquents.\n",
      "performed.\n",
      "or\n",
      "Mr,\n",
      "plaint.\n",
      "ters\n",
      "e\n",
      "constructed\n",
      "slaughterinj\n",
      "cent.\n",
      "Commerce.\n",
      "trated.\n",
      "Tfrnoney.\n",
      "factions.\n",
      "rejected.\n",
      "s\n",
      "brave.\"\n",
      "statute.\n",
      "flight\n",
      "kbrsf\n",
      "future.\n",
      ":\n",
      "oflloe.\n",
      "’08.\"\n",
      "dustries?\"\n",
      "at\n",
      "parly.\n",
      "insured.\n",
      "tluics.\n",
      "court.\n",
      "lows.”\n",
      "woman.\n",
      "7\n",
      "complaint.\n",
      "expected.\n",
      "anthem\n",
      "the\n",
      "sundown.\n",
      "same.\n",
      ".\n",
      "yen.\"\n",
      "copy.\n",
      "strong.\n",
      "Government,\n",
      "nlng.\n",
      "prloes,\n",
      "Astoria.\n",
      "?\n",
      "They\n",
      "Figaro.\n",
      "buy-\n",
      "it.\n",
      "plane.\n",
      "Nelson,\n",
      "decision.\n",
      "injunction,\n",
      "peace.\n",
      "wid\n",
      "give\n",
      "departments.\"\n",
      "costs.\n",
      "sufficient\n",
      "DOCK\n",
      "incurred.\n",
      "out\n",
      "j\n",
      "g\n",
      "his\n",
      "tho-\n",
      "mnlns.\n",
      "rder.\n",
      "trade.\n",
      "-\n",
      "Regis.\n",
      "ages.\n",
      "them.\n",
      "Mcllvalne.\n",
      "newspaper.\n",
      "matter.\n",
      "er\n",
      "improved\n",
      "be\n",
      "pastures.\n",
      "himself.\n",
      ".\n",
      "shapes.\n",
      "adopted,\n",
      "and\n",
      ":\n",
      "Robert\n",
      "the\n",
      "sheep.\n",
      "lasttwelrty8®*'\n",
      "Miss\n",
      "~\n",
      "Down\n",
      "Clerk\n",
      "circumstances.\n",
      "tomorrow's\n",
      "ncr*\n",
      "would.\n",
      "Union.\n",
      "street.\n",
      "tru.--\n",
      "n\n",
      "little\n",
      "the\n",
      "speed.\n",
      "be\n",
      "husband.,\n",
      "obtained.\n",
      "able.\n",
      "ards.\"\n",
      "situation.\n",
      "accussed\n",
      "sleep.\n",
      "«1-\n",
      "support.\n",
      "home.\n",
      "else.\n",
      "spread\n",
      "ploying.\n",
      "and\n",
      "the\n",
      "Coun-\n",
      "shot-makin-\n",
      "guarded.\n",
      "an\n",
      "Hoursfrom9a.M..to5e.m.\n",
      "Lake\n",
      "1\n",
      "rip-rapping.\n",
      "publication.\n",
      "Stanton's\n",
      "Invigorated.\n",
      "rendered.\n",
      "strike,\n",
      "cnn-\n",
      "instruction.\n",
      "territory.\n",
      "ly.\n",
      "the\n",
      "ascertained.\n",
      "service.\n",
      "unrelenting.\n",
      "York.\n",
      "to-\n",
      "pursuits.\n",
      "for­\n",
      "ever\n",
      "plates.\n",
      "Meade.\n",
      "superior\n",
      "Dusinesa.\n",
      "ho-\n",
      "locomotives.\n",
      "derson.\n",
      "like\n",
      "Flag.\n",
      "night.\n",
      "hla\n",
      "May.\n",
      "r\n",
      "Health.\n",
      "*\n",
      "term\n",
      "provinces.\n",
      "off.\n",
      "transao-\n",
      "ferine.\n",
      "the\n",
      "preserves\n",
      "father\n",
      "them.\n",
      "hopeful.\n",
      "\"\n",
      "it.\n",
      "jfciow.\n",
      "stopped.\n",
      "other,\n",
      "sentiments.\n",
      "action,\"\n",
      "harlequinade.\n",
      "water.\n",
      "\"detective.\"\n",
      "tutlon.\n",
      "institutions.(12)\n",
      "dreary.\n",
      "was\n",
      "say\n",
      "maker.\n",
      "nies.\n",
      "vagabond.\n",
      "here.\n",
      "fault.\n",
      "and\n",
      "EU.\n",
      ";\n",
      "fur.\n",
      "agreement.\n",
      "11,012\n",
      "provide.\n",
      "Sovereign.\n",
      "life.\n",
      "law.\n",
      "immeiatly\n",
      "ity.\n",
      "law.\n",
      "I\n",
      "faction.\n",
      "nificance.\n",
      "mac\n",
      "7p.m.\n",
      "James\n",
      "individuals.\n",
      "sure.\n",
      ">n»v.\n",
      "city.\n",
      "brane.\n",
      "relations.\n",
      "if\n",
      "Moka\n",
      "Press.\n",
      "during\n",
      "relatives\n",
      "colors.\n",
      "n-\n",
      "Arm.\n",
      "it-\n",
      "from\n",
      "J'e\n",
      "un-\n",
      "er\n",
      "2\"\n",
      "vote?\n",
      "disease.\n",
      "feetlothepUceof\n",
      "postponed.\n",
      "and\n",
      "JyTs\n",
      "Vlmndcaler.\n",
      "supervision.\n",
      "PANY,\n",
      "-\n",
      "ceptance:\n",
      "constituents.\n",
      "e,\n",
      "interest.\n",
      "ult.\n",
      "the\n",
      "attention.\n",
      "cans\n",
      ".\n",
      "s\n",
      "pt\n",
      "day.\n",
      "p,\n",
      "reply.\n",
      "war.\n",
      "purpose\n",
      "ann,\n",
      "Mass.\n",
      ";\n",
      "form­\n",
      "o\n",
      "d\n",
      "t:\n",
      "ou\n",
      "basis.”\n",
      "comforter.\n",
      "of.\"\n",
      "State.\n",
      "search.\n",
      "us.\n",
      "cir-\n",
      "Itnddmus,\n",
      "was\n",
      "invited.\n",
      "damageB.\n",
      "study.\n",
      "cussion.\n",
      "Afpeal-\n",
      "orders.\n",
      "the\n",
      "organ-laa-\n",
      "1\n",
      "service.\n",
      "damage.\n",
      "zens.\n",
      "Science.\n",
      "superior.\n",
      "directions.\n",
      "known\n",
      "unsuccessful.\n",
      "'\n",
      "village.\n",
      "Uw\n",
      "pay.\n",
      "In\n",
      "o’clock.\n",
      "ruin.\n",
      "mil.\n",
      "sale\n",
      "o\n",
      "simply\n",
      "the\n",
      "Sun.\n",
      "—,\n",
      "statute.\n",
      "each.\n",
      "viciously.\n",
      "Small's\n",
      "Galveston\n",
      "Gazelle\n",
      "two\n",
      "passed.\n",
      "cent.\n",
      "death.\n",
      "ex-\n",
      "proof\n",
      "low-c-\n",
      "arbitration.\n",
      "times.\n",
      "deans.\n",
      "directed.\n",
      "saloon.\n",
      "The-\n",
      "me.\n",
      "home.\n",
      "men.\n",
      "Rogers.\n",
      "oootomj\n",
      "x.\n",
      "from\n",
      "fire.\n",
      "plies.\n",
      "I,.\n",
      "ones.\n",
      "location.\n",
      "e.\n",
      "an\n",
      "sake.\n",
      "s\n",
      "shall\n",
      "fire.\n",
      "Herald,\n",
      "soldbv\n",
      "citizen.\n",
      "ordinance.\n",
      "1902,\n",
      "whose\n",
      "dry\"\n",
      "election.\n",
      "pack.\"\n",
      "side.\n",
      "course\n",
      "and\n",
      "the\n",
      "b-\n",
      "day,\n",
      "e-\n",
      "outlaws.\n",
      "roots.\n",
      "work.\n",
      "physicians.\n",
      "however.\n",
      "Kulp,\n",
      "teachings\n",
      "!)ile,\n",
      "10%<\n",
      "cemetery.\n",
      "name?’\n",
      "course.\n",
      "oiu.\n",
      "clergy.\n",
      "peace\n",
      "to\n",
      "saved.\n",
      "government.\n",
      "commonplaces.\"\n",
      "week\n",
      "meet.\n",
      "1858.\n",
      "Air«.\n",
      "Mass.,\n",
      "county.\n",
      "associations.\n",
      "up.\n",
      "eouth.\n",
      "a-\n",
      "de-\n",
      "Oroville.\n",
      "o\n",
      "tc\n",
      "evening.\n",
      "state.\n",
      "oDservor.\n",
      "witdom\n",
      "circumstances.\n",
      "snags\n",
      "shown,\n",
      "farm¬\n",
      "way\n",
      "hat\n",
      "plause.)\n",
      "on\n",
      "Grahamhes.\n",
      "opponent's\n",
      "the\n",
      "fying.\n",
      "contractor.\n",
      ".iV\"\n",
      "and\n",
      "his\n",
      "1\n",
      "brother-in-la-\n",
      "engaged.\n",
      "inspection.\n",
      "Cemetery.\n",
      "country\n",
      "tale.\n",
      "stops.\"\n",
      "rocks.\n",
      "medicine.\"\n",
      "men.\n",
      "Zipf.\n",
      "Socle'y.\n",
      "erased\n",
      "Mountains.-\n",
      "30.\n",
      "aeiil\n",
      "no\n",
      "Opinion.\n",
      "it.\n",
      "have\n",
      "Robbery.—\n",
      "Brnc#.\n",
      "discussions.\n",
      "day?\n",
      "not\n",
      "ewYork.\n",
      "measure.\n",
      "and\n",
      ".M\n",
      "president.\n",
      "prim\n",
      "drops.\n",
      "hell.\n",
      "can.\n",
      "de=\n",
      "hum.\n",
      "im-a-\n",
      "aucceesluily\n",
      "Milton\n",
      "Southwestern.\n",
      "dollars.\n",
      "poles.\n",
      "by\n",
      "Mr.\n",
      "babl\n",
      "1862.\n",
      "action.\n",
      "order.\n",
      "once.\n",
      "yield.\n",
      "in.\n",
      "l'.\n",
      "God.\n",
      "capital.\n",
      "missionaries\n",
      "plastered.\n",
      "caution.”\n",
      "No.\n",
      "matter.\n",
      "licked.\")\n",
      "body.\n",
      "votes.\n",
      "relation.\n",
      "I\n",
      "them.\n",
      ".'s.\"\n",
      "joint.\n",
      "suffrage.\n",
      "list.\n",
      "i\n",
      "i\n",
      "strengtn.\n",
      "slain.\n",
      "•l\n",
      "plaint.\n",
      "School,.\n",
      "thir\n",
      "Uepubllcnn-stcel-\n",
      "ablte.\n",
      "amendment.\n",
      "water.\n",
      "climb.\n",
      "do.\n",
      "coffee.\n",
      "impossible.\n",
      "feats.\n",
      "rights,\n",
      "detectives.\n",
      "two\n",
      "nalinns.\n",
      "home.\n",
      "le-intlon.\n",
      "It.\n",
      "weeks.\n",
      "sncrarT\n",
      "it.\n",
      "reads.\n",
      "at\n",
      "r;\n",
      "d\n",
      "hcriffs\n",
      "New\n",
      "levied.\n",
      "S.\n",
      "consideration.\n",
      "of.\n",
      "old\n",
      "inquiry.\n",
      "clerk9.\n",
      "both.\n",
      "inn-ha-\n",
      "earth.-\n",
      "of\n",
      "ace.\n",
      "Astor.\n",
      "dertaking.\n",
      "Canada.\n",
      "trim.\n",
      "qualiiica-\n",
      "case.\n",
      "If\n",
      "m\n",
      "throats.\n",
      "hero.\n",
      "infidels.\n",
      "was:\n",
      "section.\n",
      "preserves.\n",
      "mine.\n",
      "road.\n",
      "-\n",
      "<nu\n",
      "organs.\n",
      "ones.\n",
      "investigation.\n",
      "available.\n",
      ",\n",
      "begin.\n",
      "it.\n",
      "Ifui-\n",
      "section.\n",
      "wees.\n",
      "counties\n",
      "importance,\n",
      "e.\n",
      "efl\n",
      "price.\n",
      "men.\"\n",
      "the\n",
      "mer.\n",
      "large.\n",
      "ed\n",
      "f/rano\n",
      "western\n",
      "for\n",
      "people.\n",
      "dition.\n",
      "preserved.”\n",
      "theless.\"\n",
      "concerned.\n",
      "others.\n",
      "supply\n",
      "v\n",
      "each.\n",
      "oousnain,\n",
      "died.\n",
      "season.\n",
      "city.\"\n",
      "is\n",
      "Koehn.\n",
      "arrested.\n",
      "says:\n",
      "nisla-\n",
      "close.\n",
      "ning.\n",
      "made.\n",
      "business.\n",
      "followers.\n",
      "voting.)\n",
      "die.\n",
      "duty.\n",
      "has\n",
      "life.\n",
      "such\n",
      "«6«.\n",
      "Reamey.\n",
      "st\n",
      "made\n",
      "lturen.\n",
      "vicinity\n",
      "ten.\n",
      "the\n",
      "matter.\n",
      "legs.\n",
      "was\n",
      "prosecuted.\n",
      "pilgrim-\n",
      "this\n",
      "Tost.”\n",
      "river.\n",
      "ers\n",
      "nomination\n",
      "persons.\n",
      "more.\n",
      "was\n",
      "nations.\n",
      "me.\n",
      "Yale.\n",
      "section.\n",
      "ofKossuth.\n",
      "prices.\n",
      "argu-\n",
      "by\n",
      "Almy.\n",
      "Pe°Ple\n",
      "peace\n",
      "be\n",
      "In-\n",
      "up\n",
      "expenses.\n",
      "cries.\n",
      "and\n",
      "go.\n",
      "or\n",
      "by\n",
      "stain.\n",
      "nuuuci\n",
      "tilicum,\n",
      "1920.\n",
      "s\n",
      "reports\n",
      "General\n",
      "settled.\n",
      "elevation.\n",
      ".\n",
      "for\n",
      "man\n",
      "kid\n",
      "Messenger.\n",
      "prisoners.\n",
      "associ\n",
      "was\n",
      "with\n",
      "o-\n",
      "destitute.\n",
      "Ledger.\n",
      "trade.\n",
      "belonging.\n",
      "quiet.\n",
      "communities.\n",
      "hostess\n",
      "table.\n",
      "Washington.\n",
      "room.\n",
      "markettouae?\n",
      "doctrine.\n",
      "Slate.\n",
      "creditors\n",
      "curse.\n",
      "taxes.\n",
      "throat.\n",
      "janitors,\n",
      "the\n",
      "was\n",
      "company.\n",
      "Cincinnati\n",
      "Arthur-\n",
      "sufferers.\n",
      "a\n",
      "Md\n",
      "hand.\n",
      "Junl'i\n",
      "*V,«n/l\n",
      "up.\"\n",
      "ihe\n",
      "when\n",
      "aforementioned.\n",
      "work.\n",
      "air.\"\n",
      "pass,\n",
      "theaters.\n",
      "school\n",
      "tive.\"\n",
      "town.\n",
      "Philadelphia.\n",
      "ww\n",
      "Jersey.\n",
      "Del.;\n",
      "go.\n",
      "housekeeping.\n",
      "disfranchisement?\n",
      "Damascus.\n",
      "t:\n",
      "discretion.\n",
      "many\n",
      "of\n",
      "HUUNIOSf.\n",
      "progress.\n",
      "Hartford\n",
      "first.\n",
      "tenderness,\n",
      "III.\n",
      "4,300.\n",
      "and\n",
      "bullock.\n",
      "l.lvcr\n",
      "therefor.\n",
      "produced\n",
      "th;ng.\n",
      "done.”\n",
      "pbouta?\n",
      "school\n",
      "thefnii\n",
      "1883,.\n",
      "the\n",
      "Jones.\n",
      "\".Exchange.\n",
      "nothing.\n",
      "manufacture\n",
      "generations.\n",
      "163.\n",
      "j\n",
      "remedy.\n",
      "runs.\n",
      "me.\n",
      "date\n",
      "ion:\n",
      "quired.\n",
      "act.\n",
      "administered\n",
      "him.\"\n",
      "aspect\n",
      "--Moines;\n",
      "melody,\n",
      "public\n",
      "him.\n",
      "Ward.\n",
      "Ipring?\n",
      "two-thir-\n",
      "by\n",
      "him\n",
      "Gil.\n",
      "macnines.\n",
      "respectfully,\n",
      "chapeau.\n",
      "line:\n",
      "inter-\n",
      "railway\n",
      "ihw\n",
      "Journal.\n",
      "pan.\n",
      "well.\n",
      "use.\n",
      "lines.'1\n",
      "glee.\n",
      "north,\n",
      "columns.\n",
      "moment,\n",
      "deep-seate-\n",
      "i\n",
      "eighty-one\n",
      "its\n",
      "slgs.\n",
      "remedy.\n",
      "•\n",
      "week.\n",
      "Plague.\n",
      "lived.\n",
      "stop.\n",
      "conclusions.\n",
      "and\n",
      "State.\n",
      "advanta.e\n",
      "appearance.\n",
      "preserver.\n",
      "one\n",
      "sun.\n",
      "191--\n",
      "G\n",
      "board.\n",
      "r,\n",
      "season.\"\n",
      "add.\n",
      "formation.\n",
      "\"safety.\"\n",
      "one.\n",
      "Co.,\n",
      "Benja\n",
      "unjust.\n",
      "enrap-\n",
      "returned.\n",
      "tho\n",
      "howIcanwl\n",
      "Berlin.\"\n",
      "affections.\n",
      "take\n",
      "-\n",
      "In\n",
      "road.\n",
      ".\n",
      "conventions.\n",
      "hands.\n",
      "keep\n",
      "applic&tant.\n",
      "conviction.\n",
      "across.\n",
      "kind.\n",
      "due.\"\n",
      "state.\n",
      "grow.\n",
      "built,\n",
      "large\n",
      "«;nt.\n",
      "existence\n",
      "Booud.\"\n",
      "?\n",
      "Meade.*\n",
      "the\n",
      "first.\n",
      "case\n",
      "riage\n",
      "Hal\n",
      "vice.\n",
      "winter.\n",
      "potatoes.\n",
      "sca-\n",
      "necessary.\n",
      "critics.\n",
      "-\n",
      "defend-\n",
      "a\n",
      "balcony.\n",
      "rum.\n",
      "power\n",
      "Philippines.\n",
      "nomination.\n",
      "S.\"\n",
      "of\n",
      "bees.\n",
      "Police\n",
      "4t\n",
      "said.\n",
      "mental\n",
      "perceptible.”\n",
      "recommitted.\n",
      "legislation,\n",
      "palling.\n",
      "health.\n",
      "MOB.\n",
      "on\n",
      "of\n",
      ",\n",
      "the\n",
      "Constitution.\n",
      "Portsmouth.\n",
      "-\n",
      "Ilcrald.\n",
      "Uh\n",
      "ginning.\n",
      "express.”\n",
      "in\n",
      "imagination.\n",
      "Infantry;\n",
      "Blackbird.\n",
      "indulgence\n",
      "*tiy.\n",
      "oblige\n",
      "weeks.\n",
      "ma-\n",
      "4\n",
      "-\n",
      "parllsan\n",
      "troop.\n",
      "time.\"\n",
      "clothes.\n",
      "and\n",
      "hours.\n",
      "inch.\"\n",
      "Company.\n",
      "$200.\n",
      "judgment.\n",
      "n\n",
      "membership.\n",
      "than\n",
      "beacon\n",
      "o'clock.\"\n",
      "quotations.\n",
      "kepthis\n",
      "brated\n",
      "o\n",
      "8.\n",
      "elseH\n",
      "fund.\n",
      "cage.\n",
      "Gon.\n",
      "C,t\n",
      "Clerk\n",
      "town.\n",
      "him.\n",
      "House.\n",
      "reached.\n",
      "Section\n",
      "enough.”\n",
      "m.\n",
      "d.\n",
      "week.\n",
      "iV\n",
      "Virginia\n",
      "nothing.\n",
      "Pa.,\n",
      "tL\n",
      "example.\n",
      "eflprts.\n",
      "edbyi\n",
      "man's.\n",
      "liberties\n",
      "lifetime.\n",
      "St.\n",
      "emulate.\n",
      "so\n",
      "uly,\n",
      "is\n",
      "spring.\n",
      "medicine\n",
      "16^0.\n",
      "of\n",
      "Newark.\n",
      "house.\n",
      "work.\n",
      "popular.\n",
      "oent.\n",
      "Rbelms.\n",
      "madhouse.\n",
      "4\n",
      "ge.\"\n",
      "d\n",
      "again?\n",
      "portation.\n",
      "parties.\n",
      "dent.\n",
      "sensation.\n",
      "Church.\n",
      "too.\n",
      "aloresmd.\n",
      "university.\n",
      "proceedings.\n",
      "two\n",
      "company.\n",
      "ft.)\n",
      "ernor.\n",
      "North-\n",
      "C.\n",
      "\"\n",
      "ia-\n",
      "expectation.\n",
      "drawers\n",
      "Bhe\n",
      "it\n",
      "built__yet.\n",
      "trial.\n",
      "Phlludelphla.\n",
      "dollars.\n",
      "association.\n",
      "amusement\n",
      "D.li,toA.8.\n",
      "articles.\n",
      "drunken.\n",
      "cedure.\n",
      "them.\"\n",
      "iticc,\n",
      "character.\n",
      "votes.\n",
      "feet.\n",
      "point.\n",
      "she\n",
      "ty.\n",
      "n\n",
      "county.\n",
      "rs.\n",
      "plans.\n",
      "Papist.'\n",
      "Bulletin.\n",
      "World;\n",
      "deserves\n",
      "tables.\n",
      "l»Hi.\n",
      "Is\n",
      "ropulsivess.\n",
      "not\n",
      "statute.\n",
      "itanced.\n",
      "natives\n",
      "that\n",
      "fare.\n",
      "forefathers.\n",
      "-\n",
      "i\n",
      "law.\n",
      "likely.\n",
      "re.\n",
      "beginning.\n",
      "hands.\"\n",
      "peninsula.\n",
      "ber.\n",
      "650\n",
      "day.\n",
      "turpentine.\n",
      "benefac\n",
      "lOO/alOtt1,.\n",
      "x\n",
      "toastmaster.\n",
      "struggle.\n",
      "forms\n",
      "ss\n",
      "ago.\n",
      "sell.\n",
      "I\n",
      "Kronch\n",
      "work.\n",
      "faithful\n",
      "and\n",
      "out.\n",
      "a\n",
      "order.\n",
      "and\n",
      "uu.\n",
      "shock.\n",
      "things.\n",
      "daily.\n",
      "But\n",
      "of\n",
      "o\n",
      "Island.\n",
      "It.\n",
      "and\n",
      "Senatorial\n",
      "death.\n",
      "-\n",
      ",\n",
      "advance.\n",
      "lives.\n",
      "country.\"\n",
      "hi\n",
      "desire,\n",
      "applicatant.\n",
      "”\n",
      "iron\n",
      "ill-fat-\n",
      "win.\n",
      "them.\n",
      "utmost\n",
      "Governor.\n",
      "ity\n",
      "Cnesbire.\n",
      "««\n",
      "Wilmington.\n",
      "bttice.\n",
      "hearth.\n",
      "sion.\n",
      "m.\n",
      "peril\n",
      "flow­\n",
      "combination.\n",
      "set\n",
      "tilfiel?\n",
      "equullj-\n",
      "example.\n",
      "eoniMder.iltle\n",
      "Sunday.\n",
      "Taylor\n",
      "origin.\n",
      "work.\n",
      "emergency.\n",
      "of\n",
      "peck.\n",
      "dlence.\n",
      "air.\n",
      "country.\n",
      "-\n",
      "October\n",
      "volunteering.\n",
      "Elmwood.\n",
      "line.\n",
      ".\n",
      "manner\n",
      "christian\n",
      "exists.”\n",
      ",\n",
      "straw-\n",
      "-\n",
      "-i\n",
      "st\n",
      "heart.\n",
      "occasion,\n",
      ".\n",
      "Brown.\n",
      "schools.\n",
      "9.\n",
      "reins.\n",
      "caretaker\n",
      "months.\n",
      "Senate.\n",
      "ago\n",
      "d\n",
      "of\n",
      "work.\n",
      "trad.\n",
      "ed\n",
      "Journal\n",
      "world.\n",
      "Arcadia.\n",
      "instruction.\n",
      "S.\n",
      "ved.\n",
      "them.\n",
      "murderer.’\n",
      "intorests\n",
      "Democrats\n",
      "rc\n",
      "lime.\n",
      "tnblo.\n",
      "excellence.\n",
      "116th.\n",
      "whole.\"\n",
      "Orleans.\n",
      "becoming\n",
      "another.\n",
      "jahr.”\n",
      "e\n",
      "Kt»\n",
      "rules.\n",
      "*ac.\n",
      "'\n",
      "’\n",
      "lives.\n",
      "other.\n",
      "Science.\n",
      "chance.\n",
      "Catawlssa,\n",
      "aeain.\n",
      "Magazine.\n",
      "ifidvnce.\n",
      "mine.\"\n",
      "offer\n",
      "I\n",
      ".to\n",
      "loyal\n",
      "Mount\n",
      "Age.\n",
      "schedule.\n",
      "a^t«sh\n",
      "sustained.\n",
      "14th.\n",
      "drops\n",
      "communicable\n",
      "Slate.originally\n",
      "In\n",
      "t\n",
      "tiller.\n",
      "State.\n",
      "father\n",
      "effective.\n",
      "tive\n",
      "high\n",
      "e\n",
      "ernments.\n",
      "food\n",
      "rock!\n",
      "tape'\n",
      "distribution.\n",
      "same.\n",
      "Capitol.\n",
      "Iceland.\"\n",
      "afore-\n",
      "know.\n",
      "bushel\n",
      "name.\"\n",
      "new\n",
      ";\n",
      "yours,\n",
      "made.\n",
      "be¬\n",
      "State*.\n",
      "her.\n",
      "President-Wilson'-\n",
      "arrangement.\n",
      "Advertiser.\n",
      ",an-\n",
      "Schamyl’s\n",
      "selling.\"\n",
      "mentioned.\n",
      "diagram:\n",
      "list.\n",
      "character.\"\n",
      "ed\n",
      "appurtenances.\n",
      "de-\n",
      "maintenance\n",
      "time.\n",
      "legislature.\n",
      "I>*s7.\n",
      "combination.\n",
      "loco-\n",
      "elae.'\n",
      "Stat*.\n",
      "follows:\n",
      "windpipe.\n",
      "e-\n",
      "Church.\n",
      "November,\n",
      "shall\n",
      "streets.\n",
      "dlea\n",
      "d\n",
      "5:18).\n",
      "Journal.\n",
      "usual.\n",
      "remedy.\n",
      "guilt.\n",
      "the\n",
      "Jr.,\n",
      "of\n",
      "for.\n",
      "information.\n",
      "experiences.\n",
      "cerned.\n",
      "certain-.\n",
      "10,0(58.'\n",
      "Republic.\n",
      "hundred\n",
      "and\n",
      "citizens.\"\n",
      "weaken.\"\n",
      "tion.\n",
      "facte\n",
      "y,\n",
      "zation.\"\n",
      "us.\n",
      "will-trem-ble\n",
      "e\n",
      "York.\n",
      "copy.\n",
      "Jer-\n",
      "the\n",
      "Mrs.\n",
      "White.\n",
      "feet\n",
      "If\n",
      "charges.\n",
      "country.\"\n",
      "suffer.\n",
      "existence.\n",
      "flag.\n",
      "two.\n",
      "at\n",
      "act.\n",
      "city.\n",
      "rita'es,\n",
      "A\n",
      "answer.\n",
      "ner.\n",
      "camp.\n",
      "remedy.\n",
      "(13)\n",
      "trouble\n",
      "pastures.\n",
      "and\n",
      "street.\n",
      "Jersey.\n",
      "honor.\"\n",
      "decision?\n",
      "hours.\n",
      "bunker.\n",
      "neighbors.\n",
      "more.\n",
      "“\n",
      "It\n",
      "Polynesian.\n",
      "\"\n",
      "the\n",
      "around.\n",
      "uithsea.\n",
      "forovor.\"\n",
      "the\n",
      "home.\n",
      "Page's\n",
      "Jr.\n",
      "situate--\n",
      ":\n",
      "of\n",
      "llnotoTur\n",
      "are\n",
      "t.\n",
      "thus:\n",
      "breath:\n",
      "18W-\n",
      "Madeline\n",
      "States.\n",
      "davs.\n",
      "donations.\n",
      "Curtis-Wright.\n",
      "time!\"\n",
      "armies.\n",
      "hot\n",
      "terday:\n",
      "never\n",
      "treasury.\n",
      "health.\n",
      "^\n",
      "sanitariums.\n",
      "it.\"\n",
      "house.\n",
      "s\n",
      "old.\n",
      "life.\n",
      "arrest..\n",
      ".\n",
      "Co.,\n",
      "of\n",
      "iuet\n",
      "coirered.\"\n",
      "the\n",
      "friends\n",
      "order.\n",
      "t\n",
      "district,\n",
      "protection.\n",
      "pulpit\n",
      "lost.\n",
      "grown.\n",
      "of\n",
      ",k.\n",
      "Judges.\n",
      "recorder.\n",
      "Mr.\n",
      "gunpowder.\n",
      "I'arlors.\n",
      "d\n",
      "system.\n",
      "a\n",
      "horizon.\n",
      "now.\n",
      "dear.\"\n",
      "day.\n",
      "winners\n",
      "Herald.\n",
      "washes!\n",
      "full.\n",
      "blood.\n",
      "there\n",
      "leaaantly,\n",
      "plating.\n",
      "wholesale.\n",
      "wti-x\n",
      "c\n",
      "to\n",
      "and\n",
      "boost.\n",
      "wire.\n",
      "morality.\n",
      "beginning.\n",
      "cneap.\n",
      "Leroux,\n",
      "propriations.\n",
      "union.’’\n",
      "Dlspatcl.\n",
      "Conference.\n",
      "Sart\n",
      ",\n",
      "tured.\n",
      "o\n",
      "d\n",
      "-\n",
      "or\n",
      "pitable.\n",
      "uay.\n",
      "bales.\n",
      "Emanuel.\n",
      "change,\n",
      "happiness.\n",
      "e\n",
      "Pott.\n",
      "a\n",
      "male.\n",
      "see\n",
      "cf\n",
      "*1.392,180,294.\n",
      "expenses.\n",
      "h\n",
      "al\n",
      "be\n",
      "out.\n",
      "hear\n",
      "publics.\n",
      "1909.\n",
      "dusi-\n",
      "wrought.\"\n",
      "862,000.\n",
      "him.\n",
      "o\n",
      "multitude.'\n",
      "South.\n",
      "d-\n",
      "pany.\n",
      "Raleiffh\n",
      "$1t)0.\n",
      "girl.'\n",
      "days;\n",
      "feet.\n",
      "ago:\n",
      "proof.\n",
      "Union.\n",
      "of\n",
      "tests.\n",
      "injuries.\n",
      "determined.\n",
      "the\n",
      "-\n",
      "1916.\n",
      "Dr.\n",
      ",\n",
      "persons.\n",
      "each.\n",
      "bu-\n",
      "gerous?\n",
      "yesterday.\n",
      "winter.\n",
      "22\n",
      "tion,\n",
      "increases.\n",
      "fash-\n",
      "contemplate.\n",
      "needless.\n",
      ".\n",
      "J\n",
      "Preference.\n",
      "adjourned.\n",
      "per.\n",
      "by\n",
      "planets.\n",
      "turn.\n",
      "way,\n",
      "rd\n",
      "women\n",
      "the\n",
      "the\n",
      "with\n",
      "ishnient.\n",
      "it\n",
      "e\n",
      "state.\n",
      "year.\n",
      "donor.\n",
      "r\n",
      "executed.\n",
      "Judge.\n",
      "effected.\n",
      "cruisers.\n",
      "Bhreveport.\n",
      "4c.”\n",
      "to-da-\n",
      "bosom\n",
      "hers.\n",
      "us.\n",
      "renown.\n",
      "Island.\n",
      "d\n",
      "Jlouse.\n",
      "Ineversawaman\n",
      "could\n",
      "wet.\n",
      "everything's\n",
      "Louis.\n",
      "present\n",
      "returns.\n",
      ".OOO\n",
      "nil\n",
      "!...<>\n",
      "it.\n",
      "prisoner,\n",
      "1903.\n",
      "Class.”\n",
      "...\n",
      "Weekly.\n",
      "made\n",
      "other\n",
      "outstretched\n",
      "$43,000\n",
      "case.\n",
      ":\n",
      "rice.\"\n",
      "ob|ec?.CS.,°\n",
      "American.\n",
      "(4)\n",
      "dinner.\n",
      "place.\n",
      "tive.\n",
      "ud-\n",
      "months.\n",
      "days.\n",
      "place.\n",
      "accession\n",
      "to\n",
      "boUhcvism\n",
      "friend.\n",
      "re-union.\n",
      "ter\n",
      "street.\n",
      "1938.\n",
      "hits.\n",
      "29.\n",
      "comparison.\n",
      "is.\n",
      "[\n",
      "“\n",
      "demonstrate.\n",
      "thereto.\n",
      "offspring.\n",
      "the-\n",
      "1\n",
      "most\n",
      "said\n",
      "colonics.\n",
      "lives.\n",
      "tively.\n",
      "sure.\n",
      "sonslderatlon\n",
      "committed.\n",
      "witli\n",
      "transport.\n",
      "(Nev.)Enterprise.\n",
      "BARNABE.\n",
      "discharged.\n",
      "to\n",
      "e,\n",
      "nearly\n",
      "t\n",
      "for.\n",
      "cash.\n",
      "nr-iK-\n",
      "giave.\n",
      "wifcrpoi\n",
      "*eBult.\n",
      "pose.\n",
      "sun.\n",
      "spakc:\n",
      "remain\n",
      "rights.\n",
      "if\n",
      "comment.\n",
      "attention.\n",
      "enormou\n",
      "2.'\n",
      "woodman\n",
      "servation.\n",
      "Matter.\n",
      "Md.\n",
      "hand.\n",
      "cago,\n",
      "desired.\n",
      "so\n",
      "County\n",
      "best\n",
      ".\n",
      "'\n",
      "affair.\n",
      "confidence.\n",
      "accepted.\n",
      "sight.\n",
      "settled.\n",
      "way.\n",
      "Missouri.\n",
      "r,\n",
      "ordinances\n",
      "1\n",
      "avenue.\n",
      "isalts.\n",
      ".\n",
      "direction.\n",
      "trucks.\n",
      "»*>\n",
      "In\n",
      "John.\n",
      "Department.\n",
      "oyes.\n",
      "?,.\n",
      "as\n",
      "l.s\n",
      "-I\n",
      "worth\n",
      "promised.\n",
      "hum-dru-\n",
      "man.\n",
      "endorsed\n",
      "J.,\n",
      "b\n",
      "Mich.\n",
      "eater.\n",
      "welfare.\n",
      "misery\n",
      "i«*\n",
      "game.\n",
      "reduced.\n",
      "Moses.\n",
      "future.\n",
      "advance?\n",
      "service,\n",
      "crops:\n",
      "t\n",
      "procured.\n",
      "pendent.\n",
      "making.\n",
      "unknown.\n",
      "market.\n",
      "Theatre.\n",
      "d\n",
      "M.\n",
      "settlement:\n",
      "Press.\n",
      "IMS.\n",
      "mass\n",
      "'localV*\n",
      "mentioned.\n",
      "73%c.\n",
      "those.\n",
      "today.\n",
      "school\n",
      "agriculture,\n",
      "unwise.\n",
      "one.'\n",
      "millions.\n",
      "made.\n",
      "decisively.\n",
      "earth.\n",
      "thereor,\n",
      "Atkin«m\n",
      "precisely\n",
      ".\n",
      "inspection.\n",
      "Pl'tff\n",
      "people.\n",
      "death.\n",
      "States.\n",
      "law;'\n",
      "Herald:\n",
      "respect.\n",
      ".\n",
      "Italian.\n",
      "lot*.\n",
      "coal\n",
      "of\n",
      "attendant.\n",
      "rule.\n",
      "\"nam?.\"\n",
      "basket.\"\n",
      "program.\n",
      "said:\n",
      "acquired\n",
      "Journal.\n",
      "M\n",
      "dead.\n",
      "$2;\n",
      "done.\n",
      "company.\n",
      "ment,\n",
      "WITH\n",
      "ased\n",
      "through\n",
      "clubs.\n",
      "wave\n",
      "18\n",
      "for\n",
      "point.\n",
      "S5.ni;:.\n",
      "views.\n",
      "bill.\n",
      "evi-\n",
      "again.\n",
      "s-\n",
      "work.\n",
      "I'm\"\n",
      "Congress.\n",
      "^\n",
      "ity.\"\n",
      "ginner.\"\n",
      "r\n",
      "place.\n",
      "witnesses,\n",
      "church-\n",
      "ers.\n",
      "head.\n",
      "J.,\n",
      ".,\n",
      "-\n",
      "men.\n",
      "thoughtful*\n",
      "UM,Ul«a>\n",
      "deficit.\n",
      "amount\n",
      "parade.\n",
      "move.\"\n",
      "Democrat.\n",
      "and\n",
      "ion.\n",
      "surgery.\n",
      "crows.\n",
      "to.\n",
      "in\n",
      "Kansas.\n",
      "Roy\n",
      "remedy.\n",
      "act.\n",
      "shall\n",
      "strikes.\n",
      "continued.\n",
      "table.\n",
      "be.\"\n",
      "navy.\n",
      "a\n",
      "ieed\n",
      "tlierefbro,\n",
      "of\n",
      "to.\n",
      "system.\n",
      "ernoon.\n",
      "Interests.\n",
      "basis.\n",
      "d\n",
      "there.\n",
      "tion.\n",
      "another.\n",
      "Advt.\n",
      "trains.\n",
      "a!\n",
      "passenger.\n",
      "Commlaaloner*.\n",
      "lor-.\n",
      "concolved.\n",
      "shirt\n",
      "uncover\n",
      "district\n",
      "energy,\n",
      "perience\n",
      "Shortell.\n",
      "prospect.\"\n",
      "Monday.\n",
      "once.\n",
      "forcements.\n",
      "Allegheny.\n",
      "$5.\n",
      "1919.\n",
      "-rupte-\n",
      "«.\n",
      "Illinois.\n",
      "notea,\n",
      "quarrels\n",
      "Wade.\n",
      "subject\n",
      "Mrs\n",
      "Acosta\n",
      "spoliation.\n",
      "ble.\n",
      "g\n",
      "~\n",
      "-\n",
      "yours.\n",
      "niece.\n",
      "umbrella.\n",
      "dians.\n",
      "return.\n",
      "camps.\n",
      "ownercl\n",
      "dCalh*\n",
      "scheme.\n",
      "stadium.\n",
      "Congress.\n",
      "fwi«i.\n",
      "travelingaisen-al-\n",
      "W.\n",
      "survives.\n",
      "dur\n",
      "u\n",
      "story.\n",
      "tent\n",
      "would\n",
      "duPont.\n",
      "trouble.\n",
      "stealing.\n",
      "ago.\n",
      "days.\n",
      "regiment.\n",
      ".'i?\n",
      "others.\n",
      "taw.\n",
      "us.\n",
      "'\n",
      "right.\n",
      "oramanheIs.\n",
      "fatal.\n",
      "military.\n",
      "which\n",
      "hymyownhand.\n",
      "debt.\n",
      "right.\n",
      "kbUL\n",
      "absurdity.\n",
      ".\"\n",
      "him.\n",
      "Central.\n",
      "soil.\n",
      "ended.\n",
      "passatrc.\n",
      ".\n",
      "done.\n",
      "d\n",
      "price,\n",
      "urgent.\n",
      "then\n",
      "town!\n",
      "S3.1\n",
      "water.\n",
      "reality.\n",
      "-\n",
      "country.\n",
      "n-\n",
      "Saulnier,\n",
      ".\n",
      "descending.\n",
      "rtcnoaswd.\n",
      "select.\n",
      "praise.\n",
      "expccta-\n",
      "and.\n",
      "proceed\n",
      "s\n",
      "extension.\n",
      "jubilee.\n",
      "lilteen.\n",
      "counter.\n",
      "one.\n",
      "importance.\n",
      ".\n",
      "street.\n",
      "years.\n",
      "Hospital.\n",
      "public,\n",
      "Atlantic.\n",
      "year*.\n",
      "master:\n",
      "ll'lSe.\n",
      "UUO.UUU.\n",
      "*9.\n",
      "Mink.\n",
      "\"\n",
      "s\n",
      "escaped.\n",
      "PhoA\n",
      "you.\"\n",
      "covers.\n",
      "flowers.\n",
      "facilities.\"\n",
      "rule.\"\n",
      "facts.\n",
      "tho\n",
      "1891,\n",
      "anxiety.\n",
      "to\n",
      "com\n",
      ".\n",
      "tbe\n",
      "appeal.\n",
      "the\n",
      "Me.\"\n",
      "rnther\n",
      "livelihood.\n",
      "?\"\n",
      "floors.\n",
      "IL\n",
      "instances.\n",
      "Moor.”\n",
      "Detroit.\n",
      "on.\n",
      "uri\n",
      "hi\n",
      "colored\n",
      "S.\n",
      "pavement.\n",
      "cneap.\n",
      "work.\n",
      "river.\n",
      "confusion.\n",
      "buildings.\n",
      "Angnstly-beautiful-\n",
      "cants.\n",
      "d\n",
      "iibtrology.\n",
      "viz.:\n",
      "distribution\n",
      "wherein\n",
      "June]!).\n",
      "y\n",
      "missioner.\n",
      "command.\n",
      "greet-\n",
      "’’\n",
      "-\n",
      "Auditorium.\n",
      "on\n",
      "bler.\n",
      "Cultivator.\n",
      "they\n",
      "letter\n",
      "northwest.\n",
      "ment.\n",
      "-\n",
      "quie\n",
      "him.\n",
      "them.\n",
      "raised.\n",
      "shaft.\n",
      "it\n",
      "therein.\n",
      "can\n",
      "Agent,\n",
      "denizen.\n",
      "tablished.\n",
      "v\n",
      "purchased,\n",
      "him.\n",
      "course.\"\n",
      "war\n",
      "kissed.\n",
      "ing:\n",
      "may\n",
      "this\n",
      "effort.\n",
      "capstone.\n",
      "minister.\n",
      "laud.\n",
      "future.\n",
      "doomed.\n",
      "philosopher.\n",
      "roads.\n",
      "people.\n",
      "debt.\"\n",
      "although\n",
      "demanded\n",
      "hands.\n",
      "over,\n",
      "r\n",
      "quired,-'\n",
      "Fitzsimmons-Pritchar-\n",
      "i?j,d.\n",
      "remedy.\n",
      ".\n",
      "growth.\n",
      "as\n",
      "Sr.u.\n",
      ".\n",
      "(lellglilcd\n",
      ".\n",
      "triple.\n",
      "Sangston.\n",
      "per-\n",
      "it\n",
      "g\n",
      "driven\n",
      "seen.\n",
      "diaries.\n",
      "corporations.\n",
      "them.\n",
      "ments.\n",
      "postpaid.\n",
      "tions.\n",
      "extermination.\n",
      "drowned.\n",
      "t\n",
      "gallows.\n",
      "to\n",
      "pan.\n",
      "financed.\n",
      "Century.\n",
      "I\n",
      "themeans\n",
      "-\n",
      "m.\n",
      "o'clock.\n",
      "1802.\n",
      "inaugural.\n",
      "utlons.\n",
      "be\n",
      "y.\n",
      "effect\n",
      "•v;.\n",
      "o\n",
      "Church.\n",
      "Smith.\n",
      "belt.\n",
      "sailor.,\n",
      "Medlfer-\n",
      "again.\n",
      "both.\n",
      "cemetery.\n",
      "time.\n",
      "tbe\n",
      "rules.\n",
      "them.\n",
      "groceries.\n",
      "c\n",
      "me.\n",
      "umphantly.\n",
      "sunbeam.\n",
      "DallaaCo.,\n",
      "officers\n",
      "covered.\n",
      "laws.\n",
      "risen.\n",
      "within.\n",
      "rnment.\n",
      "best.\n",
      "MARTYR\n",
      "which\n",
      "supply.\n",
      "road.\n",
      "vain.”\n",
      "good.\n",
      "Nar-I\n",
      "Sts.\n",
      "was\n",
      "second.\n",
      "available.\n",
      "evening.\n",
      "uway.\"\n",
      "Kitty!\n",
      "one.\n",
      "55\n",
      "Incumbents.\n",
      "of\n",
      "girl.\n",
      "fertilizer.\n",
      "people\n",
      "interior.\n",
      "95.\n",
      "pair,\n",
      "sons\n",
      "sweat,\n",
      "e\n",
      "lf\n",
      "therein.\n",
      "it\n",
      "to\n",
      "gentleman\n",
      "it.\n",
      "bottle.\n",
      "officials.\n",
      "question.\n",
      "waters.\n",
      "State.\n",
      "laws.\n",
      "Herald\n",
      "had\n",
      "West.\n",
      "work\n",
      "Vermonler.\n",
      "resolution:\n",
      "desirable.\n",
      "21st.\n",
      "Lincoln\n",
      "Store.\n",
      "months.\n",
      "ordinance.\n",
      "school.\n",
      "It\n",
      "state\n",
      "Wilmington.\n",
      "cnmpcm-ntioi-\n",
      "possoselon.\n",
      "11\n",
      "l'\n",
      "classes.\n",
      "41\n",
      "occurrence.\n",
      "breeze.\n",
      "war.\n",
      "purposes.\n",
      "aim.\n",
      "ness.\n",
      "{140,000,000\n",
      "reply.\n",
      "birth.\n",
      "par\n",
      "enduring.'\"\n",
      "out\n",
      "finished.\n",
      "but\n",
      "which\n",
      "barrel.\n",
      "teams.\n",
      "s.nVI\n",
      "the\n",
      "complaint.\n",
      "sorrow\n",
      "Hoffecker.\n",
      "lit.\n",
      "by\n",
      "was\n",
      "possible.\n",
      "safe-\n",
      "instructions.\n",
      "afterward.\n",
      "seemed\n",
      "about.\n",
      "serious.\n",
      "fighting.\n",
      "deemed,\n",
      "ages.\n",
      "rule.\n",
      "1860.\n",
      "Jdn.\n",
      "the\n",
      "Hadlej.\n",
      "ng\n",
      "idea.\n",
      "interest.\n",
      "beginning.\n",
      "remain.\n",
      "vices.\n",
      "perhaps,\n",
      "bankers.\n",
      "1)\n",
      "extension.\n",
      "tight.\"\n",
      "only.\n",
      "“\n",
      "didate.\n",
      "th\n",
      "son*.\n",
      "paper.\n",
      "his\n",
      "music.\n",
      "disin-teime-\n",
      "his\n",
      "re\n",
      "England.\n",
      "route.\n",
      "[Alta.\n",
      "ordinance.\n",
      "grade.\n",
      "t\n",
      "round.\n",
      "Leonardt.\n",
      "mischief,\n",
      "grave.\n",
      "$24.\n",
      "merchandise.\n",
      "action.\n",
      "aforesaid.\n",
      "Century.\n",
      "did\n",
      "Impies-slvo-\n",
      "yet.\n",
      "faith.\n",
      "ton.\"\n",
      "October\n",
      "digestion.\n",
      "death.\n",
      "method.\n",
      "waitress.\n",
      "offences.\n",
      "the\n",
      "nour.\"\n",
      "Rapublicaoa\n",
      ":\n",
      "Navy.\n",
      "appear\n",
      "vantages\n",
      "them.\n",
      "for\n",
      "policy.\n",
      "contract.”\n",
      "adjourn.\n",
      "crime.\n",
      "a\n",
      "speculators.\n",
      "Cooper.\n",
      "work.\n",
      "shrine.\n",
      "1877.\n",
      "construction,\n",
      "represent.\n",
      "night.\n",
      "damaged.\n",
      "Christ.\"\n",
      "chairman\n",
      "stolen\n",
      "~a.\n",
      "past\n",
      "delivery.\n",
      "wound.\n",
      "otherwise.\n",
      "prosper\n",
      "examination.\n",
      "Hitters.\n",
      "instruction.\n",
      "fsalad-\n",
      "remedy.\n",
      "friends\n",
      "World,\n",
      "nutritious\n",
      "Chapter,\n",
      "war.\"\n",
      "o\n",
      "t'\n",
      "certlflcales.\n",
      "a\n",
      "issued\n",
      "wildcats.\n",
      "advertising\n",
      "century\n",
      "school.\n",
      "er.\n",
      "an\n",
      "Court.\n",
      "thither.\n",
      "that\n",
      "politicians\n",
      "understood.\n",
      "Brigham,\n",
      "^ashiiiKtou.\n",
      "(SEAL)\n",
      "raa-\n",
      "fort.\n",
      "things.\"\n",
      "old.\n",
      "rates.\n",
      "line.\"\n",
      "dcMdtw\n",
      "rule.\n",
      "stories.\n",
      "Issue.\n",
      "I\n",
      "barbarism.\"\n",
      "Montgomery\n",
      "...\n",
      ".\n",
      "Home.\n",
      "Congress,\n",
      "esteem,\n",
      "amirnach.\n",
      "help.\n",
      "nature\n",
      "seems\n",
      "cause.\n",
      "Barker\n",
      "i\n",
      "the\n",
      "occasion,\n",
      "g\n",
      "street;\n",
      "ment:\n",
      "death.\n",
      "ladies!\n",
      "men.\n",
      "it\n",
      "above.\n",
      "is\n",
      "so.\n",
      "twlE.\n",
      "brances.\n",
      "Advocate.\n",
      "simulation.\n",
      "\"\n",
      "e\n",
      "?\n",
      "I\n",
      "canal.\n",
      "blis\n",
      "here.\n",
      "Tuesday.\n",
      "tory.\n",
      "be\n",
      "appoint­\n",
      "has-\n",
      "curds.”\n",
      "author­\n",
      "tim.\n",
      "(¦mis\n",
      "financially.\n",
      "bard.,\n",
      "order,\n",
      "days.\n",
      "\"\n",
      "cor-\n",
      "n\n",
      "fo\n",
      "tooth\n",
      "common\n",
      "edy.\"\n",
      "r\n",
      "exit.\n",
      "be\n",
      "Senate.\n",
      ".\n",
      "2\n",
      "pose.\n",
      "out.'\n",
      "Congress.\n",
      "Glory.\"\n",
      "Chronicle\n",
      "It\n",
      "to\n",
      "Sev-\n",
      "them.\n",
      "wrench.\n",
      "leisure.\n",
      "be.\n",
      "evidences.\n",
      "navy.\n",
      "to-day.\n",
      "gTMtlT.\n",
      "dissipation.\n",
      "date\n",
      "m.\n",
      "cirtificates\n",
      "weal\n",
      "west.\n",
      "Appomattox.\n",
      "miles.\n",
      "utility.\n",
      "up.\n",
      "j\n",
      "canyon.\n",
      "error.\n",
      "left.\n",
      "Bolshevik\n",
      "Transportation,\n",
      "tho\n",
      "living.\n",
      "trial.\n",
      "k-.\n",
      "go\n",
      "bounded.\n",
      "churches.\n",
      "o’clock.\n",
      "resolvents.\n",
      "the\n",
      "Fislihigcrock..-\n",
      "L\n",
      "ISIMI\n",
      "labor.\n",
      "said:\n",
      "aid.\n",
      "order.\n",
      "enormously.\n",
      "reason.\n",
      "year.\n",
      "pos-\n",
      "\\\\\n",
      "mnn.\n",
      "tinues.\n",
      "board\n",
      "mili\n",
      "ones.\"\n",
      "acquaintance\n",
      "promised.\n",
      "parts.\n",
      "Tnecranwra\n",
      "bidder.\n",
      "vigor.\n",
      "write.\n",
      "property.\n",
      "action.\n",
      "suit.\n",
      "gentlemen,\n",
      "by.”\n",
      "the\n",
      "was,-\n",
      "of\n",
      "Agnes,\n",
      "Racine.\n",
      "Mamer.\n",
      "heavy,\n",
      "eto.\n",
      "storv.\n",
      "b\n",
      "fitness.\n",
      "College,\n",
      "Intellectuals\n",
      "maximum\n",
      "man.\n",
      "lasts.\n",
      "the\n",
      "from.\"\n",
      "*\n",
      "program.\n",
      "cause.\n",
      "fering.\n",
      "Sun.\n",
      "two\n",
      "that\n",
      "givci\n",
      "carried\n",
      "Adams\n",
      "-\n",
      "faces,\n",
      "meeting.\n",
      "boats.\n",
      "installments\n",
      "vantage.\"\n",
      "tance.\n",
      "her.\n",
      "this\n",
      "here.\n",
      "in*.\"\n",
      "Gazette.\n",
      "s\n",
      "it\n",
      "meeting.\n",
      "rooster\n",
      "brick.\"\n",
      "Agent,\n",
      "World.\n",
      "the\n",
      "flesh.\"\n",
      "Mifflin.\n",
      "con-\n",
      "speedily.\n",
      "also.\n",
      "etubjrrae\n",
      "months.\n",
      "1912,\n",
      "groups.\n",
      "them.\n",
      "were\n",
      "gov-\n",
      "dent.\n",
      "charges.\n",
      "like.\n",
      "bo-\n",
      "acre.\n",
      "1866.\n",
      "painful.\n",
      "domination.\n",
      "county.\n",
      "this\n",
      "admirably.\n",
      "work\n",
      "Graphic.\n",
      "tion.\n",
      "cure.\n",
      "be\n",
      "ner.\n",
      "y-tw-\n",
      "employed.\n",
      "here.\"\n",
      "God.\n",
      "»\n",
      "means.\n",
      "testation.\n",
      "that?\n",
      "noon.\n",
      "games.\n",
      "years.\n",
      "(14)\n",
      "crat.\n",
      "jail.\n",
      "both.\n",
      "o\n",
      "re—\n",
      "were\n",
      "!?\".\n",
      "therefor.\n",
      "fruitfulness.\n",
      "nitnii,\n",
      "shrritf.\n",
      "Addlcks.\n",
      "dol­\n",
      "DEPARTMENT.\n",
      "Job,\n",
      "December,\n",
      "a\n",
      "despots,\n",
      "Times.\n",
      "time.\n",
      "iMimiitfli\n",
      "country,\n",
      "museum.\n",
      "reform.\n",
      "t:\n",
      "brother.\n",
      "ence.\n",
      "come.\n",
      "(hem.\n",
      "of\n",
      "tervenes.\n",
      "bins.\n",
      "by\n",
      "employed\n",
      "laborer\n",
      "January\n",
      "north.\n",
      "ored.\n",
      "ores.\n",
      "church.\n",
      "bht\n",
      "Palmer.\n",
      "which\n",
      "Sn\n",
      "that\n",
      "tales.\n",
      "facilities.\"\n",
      "times\n",
      "environment.\n",
      "tiser.\n",
      "earth.\n",
      "wl»?l«\n",
      "emoluments.\n",
      "which\n",
      "sluice-\n",
      "will\n",
      "l\n",
      "cop).\n",
      "attendance.\"\n",
      "fruits.\n",
      "by\n",
      "Conkllng.\n",
      "noise.\n",
      "croam.\n",
      "political\n",
      "de\n",
      "sttat-\n",
      "restaurant\n",
      "which\n",
      "him.\n",
      "clear.\n",
      "them.\n",
      "test.\n",
      "the\n",
      "Kihn.\"\n",
      "o\n",
      "overwhelming.\n",
      "streets.\n",
      "Brown.\n",
      "for\n",
      ".\n",
      "1917.\n",
      "r\n",
      "matter.\n",
      "school.\n",
      "now\n",
      "'\n",
      "merit.;\n",
      "bettor.\n",
      "hid.\n",
      "officers\n",
      "tomorrow\n",
      "the\n",
      "the\n",
      "country.\n",
      "demand.-\n",
      "States.\n",
      ";\n",
      "structure.\n",
      "p.\n",
      "bliss.\n",
      "ately.\n",
      "de\n",
      ",\n",
      "the\n",
      "camp.\n",
      "$3,000,000.\n",
      "business\n",
      "Tribune:\n",
      "cp.\n",
      "city.\n",
      "cure.\n",
      "Jury.\n",
      "roundings.\n",
      "me.\n",
      "duty.\n",
      "politics.\"\n",
      "putation\n",
      "Max.\"\n",
      "this.\n",
      "interest\n",
      "-\n",
      "'\n",
      "cattle.\n",
      "all.\n",
      "canal.\n",
      "public\n",
      "T.\n",
      "*«\n",
      "time.\n",
      "in\n",
      "aver\n",
      "ruin.\n",
      "I\n",
      "and\n",
      "them.\n",
      "?\n",
      "subject.\n",
      "of\n",
      "In.”\n",
      "ramble.\"\n",
      "d\n",
      "mended,\n",
      "explained.\n",
      "husking.\n",
      "rA\n",
      "much.\n",
      "destroyed.\n",
      "the\n",
      "-\n",
      "many.\n",
      "trw-Dr-\n",
      "tnauuer.\n",
      ".A»i,n\n",
      "M.\n",
      "delightful.\n",
      "in\n",
      "their-merr-\n",
      "July,\n",
      "-\n",
      "them.\"\n",
      "Mas­\n",
      "tonishing.\n",
      "uratiou.\"\n",
      "erally.\n",
      "values.\n",
      "theoountry.\n",
      "right.\n",
      "thereof.\n",
      "dollars.\n",
      "house.\n",
      "Josefa.\n",
      "adopted.\n",
      "elimination\n",
      "Makpm,\n",
      "b\n",
      "morning\n",
      "enter?\n",
      "s\n",
      "which\n",
      "Philadelphia.\n",
      "prosperity.\n",
      "carried.out.\n",
      "waru.\n",
      "there.\n",
      "1891,\n",
      "4.\n",
      "Congress\n",
      "throughout\n",
      "whisper.\n",
      "developments\n",
      "country.\n",
      "con-\n",
      "cc\n",
      "into\n",
      "consumption.\n",
      "you.\n",
      "time.\n",
      "said\n",
      "pound.\n",
      "disorganized\n",
      "o'clock.\n",
      "taxed.\n",
      "companions\n",
      "success\n",
      "means.\n",
      "so\n",
      "abadauLt.\n",
      "State.\n",
      "excursions.\n",
      "All\n",
      "band.\n",
      "reside.\n",
      "doctors.\"\n",
      "management.\n",
      "and\n",
      "health.\"\n",
      "haven.\n",
      "d\n",
      "operation.\n",
      "suit.\n",
      "with\n",
      "and\n",
      "system,\n",
      "skill*.\n",
      "office.\n",
      "law.\n",
      "A\n",
      "rope.\n",
      "Leady.\n",
      "spent.\n",
      "compared\n",
      "a\n",
      "most\n",
      "--\n",
      "surface.\n",
      "-\n",
      "'wife.\n",
      "assumption.\n",
      "instances.”\n",
      "ight.\"\n",
      "favor.\n",
      "advance.\n",
      "t\n",
      "fering.\n",
      "exercises.\n",
      "detstood.\n",
      "hand.\n",
      "Britain\n",
      "tai\n",
      "cneap.\n",
      "grain.\n",
      "nt\n",
      "»\n",
      "iine.\n",
      "rest.\n",
      "J24-3t\n",
      "permit.\n",
      "avenue:\n",
      "Hon.\n",
      "inink\n",
      "fast.\n",
      "tor.”\n",
      "was\n",
      "action.\n",
      "Ion.\n",
      "follows:\n",
      "infamy.\n",
      "exceptions,\n",
      "board.\n",
      "Ac,\n",
      "fession\n",
      "Fairmont.\n",
      "hearings.\n",
      "speculation.\n",
      "Lords.\n",
      "bridge.\n",
      "manner.'\n",
      "signature.\n",
      "by\n",
      "will.\n",
      "W\n",
      "Borough.\n",
      "re-\n",
      "en\n",
      "prepared\n",
      "Europe\n",
      "Commonwealth.\n",
      "examina-\n",
      "Shafer.\n",
      "hungry.\n",
      "UIUU\n",
      "1\n",
      "con-\n",
      "pleased.\n",
      "bad.\n",
      "affair.\"\n",
      "destroyed.”\n",
      "dealings.\n",
      "cess.\n",
      "bouse.\n",
      "Garrett,\n",
      "«1.782\n",
      "imported.\n",
      "f-\n",
      "sending-\n",
      ",\n",
      "granted.\n",
      "society.\n",
      "applause.]\n",
      "river.\n",
      "assignable.\n",
      "Advocate.\n",
      "field.”\n",
      "Hamburg.\n",
      "classified.\n",
      "advantage.\n",
      "the\n",
      "Union.\n",
      ".\"\n",
      "witness-\n",
      "session.\n",
      "sleep.\n",
      "prelcrred,\n",
      "brain.\n",
      "list.\n",
      "army.\n",
      "map.\n",
      "them.\"\n",
      "details.\n",
      "htar-in-\n",
      "town.\n",
      "hue.\"\n",
      "was\n",
      "doubt.\n",
      "painter.\n",
      "6.\n",
      "said,\n",
      "Basin.\n",
      "G:iz'tle\n",
      "tho\n",
      "appears-\n",
      "1*.M*\n",
      "later.\n",
      "aim.\n",
      "safe.\"\n",
      "Weekly.\n",
      "sense.\n",
      "N.,\n",
      "work.\n",
      "circuinstan\n",
      "Ridge,\n",
      "few.\n",
      "100,000.\n",
      "they\n",
      "substitute.\n",
      "divorce.\n",
      "Paper.\n",
      "Piepenbrtng,\n",
      "lorever.\n",
      "needy.\n",
      "Russia.\n",
      "\"Five-Twenty-\n",
      "•\n",
      "music.\n",
      "Ataine.)\n",
      "possible.\n",
      "hint.\n",
      "work.\n",
      "column.\n",
      "S.\n",
      "accounts,\n",
      "tho\n",
      "of\n",
      "penalty.\n",
      "old\n",
      "two-thir-\n",
      "to-day.\n",
      "shrine.\n",
      "believing.\n",
      "rapidly.\n",
      "eternity\n",
      "Breckinridge.\n",
      "wide.\n",
      "papers\n",
      "representative.\n",
      "[Oheera.]\n",
      "young.\"\n",
      "C\n",
      "documents.\n",
      "meet-\n",
      "s.\n",
      "k\n",
      "co.,\n",
      "If\n",
      "r\n",
      "action.\n",
      "veranda.\n",
      "—\n",
      "well.\n",
      "power.\n",
      "teresting.\n",
      "nights.\n",
      "1930.\n",
      "fair-mind-\n",
      "about\n",
      "\"\n",
      "lUtnamlnl,\n",
      "to.\n",
      "hai\n",
      "on.\n",
      "14%.:\n",
      "atockhold\n",
      "lives.\n",
      "Inailments\n",
      "so\n",
      "consumption.\n",
      "said\n",
      "Services.\"\n",
      "Polk!\n",
      "a\n",
      "appearance.\n",
      "scene.\"\n",
      "-,\n",
      "llelknap.\n",
      "1\n",
      "cash.\n",
      "prosperous.\n",
      "g\n",
      "superintendent's\n",
      "stars.\n",
      "county.\n",
      "the\n",
      "Ami\n",
      "ven\n",
      "Htraid.\n",
      "time.\"\n",
      "Unite\n",
      "so.\n",
      "\"\n",
      "hibited.\n",
      "\"\"\n",
      "M.\n",
      "\"\n",
      "bosses.\n",
      "that\n",
      "d\n",
      "settees,\n",
      "American.\n",
      "given.\n",
      "sight.\n",
      "the\n",
      "hearing.\n",
      "Skc.\n",
      "w?»t-\n",
      "Coinylaiut.\n",
      "government.\n",
      "great\n",
      "triumphani.\n",
      "friends.\n",
      "minutes.\n",
      "notes.\n",
      "pelerfcy.\n",
      "rearino-\n",
      "me:\n",
      "hv\n",
      "1012\n",
      "fall.\n",
      "arooad\n",
      "work\n",
      "Bradley.\n",
      "wel\n",
      "fire.\n",
      "people.\n",
      "finely.\n",
      "-\n",
      "follows:\n",
      "vandal.\n",
      "time.\n",
      "place.\"\n",
      "knowledge.\n",
      "woman\n",
      "thence\n",
      "t\n",
      "it\n",
      "seven.\n",
      "prices.\n",
      "prevail.\n",
      "a\n",
      "toilow.\n",
      "wine.\n",
      "thorn.\n",
      "the\n",
      "whereabouts.\n",
      "pasted\n",
      "2S°-.,\n",
      "impregnable.\n",
      "then.\n",
      "-\n",
      "widt\n",
      "d\n",
      "creditable\n",
      "agent.\n",
      "deavor.\n",
      "collected.\n",
      "severely.\n",
      "forthcoming.\n",
      "appear.\n",
      "a\n",
      "practice.\n",
      "you.\n",
      "brances.\n",
      "Kenuett\n",
      "1912.\n",
      "Union.\"\n",
      "possible.\n",
      "ComImloner*.\n",
      "«fcc.\n",
      "was\n",
      "advertisement\n",
      "pig-\n",
      "light.\n",
      "M.\n",
      "work.\n",
      "stone.\n",
      "bjdrug-\n",
      "Clerk\n",
      "Eagle,\n",
      "Messrs.\n",
      "000.\n",
      "largest\n",
      "night.\n",
      "-\n",
      "passed.\n",
      "eagerly\n",
      "e\n",
      "required.\n",
      "k\n",
      "Castle.\n",
      "$3.25?\n",
      "such\n",
      "possibilities,\n",
      "Journal.\n",
      "elevation.\n",
      "renovate,\n",
      "the\n",
      "Tho\n",
      ".\n",
      "the\n",
      "cemetery\n",
      "n-\n",
      "the\n",
      "ard.\"\n",
      "beyond?\n",
      ".\n",
      ".\n",
      "thm\n",
      "most\n",
      "feet.\n",
      "cheek.\n",
      "known.\n",
      "n\n",
      "useless,\n",
      "Marr\n",
      ",7\n",
      "Tribune\n",
      "over.\n",
      "slightly\n",
      "Saturday\n",
      "execution.\n",
      "burn-\n",
      "N.\n",
      "autmal\n",
      "holes.\n",
      "acted,\n",
      "Oregon.\n",
      "servation.\n",
      "be-\n",
      "prevent\n",
      "sort.\n",
      "Mary\n",
      "offer.\n",
      "feet.\n",
      ".\n",
      "bushel.\"\n",
      "for\n",
      "Montgomery\n",
      "pered:\n",
      "he\n",
      "Wheeling,\n",
      "co-owner\n",
      ":\n",
      "railway.\n",
      "es\n",
      "gage.\n",
      "il\n",
      "(father-in-la-\n",
      "Gundelflnger.\n",
      "III.\n",
      "ng\n",
      "our\n",
      "Jlotmil.\n",
      "admiration.\n",
      "him.”\n",
      "thorn\n",
      "professions.\n",
      "felt.\n",
      "¬\n",
      "ih»\n",
      "out-\n",
      "her,\n",
      "jecture.\n",
      "service.\n",
      "wood.\"\n",
      "resort.\n",
      "Butt**\n",
      "shrine.\n",
      "Instruction,\n",
      "f\n",
      "non.\n",
      "up.”\n",
      "WTIC,\n",
      "sec.”\n",
      "rt\n",
      "flora.\"\n",
      "business.\n",
      "s\n",
      "for\n",
      "Hour*-\n",
      "proteet\n",
      "little\n",
      "cheated\n",
      "Result\n",
      "»ked.\n",
      "Telegram.\n",
      "should\n",
      "it\n",
      "to8p.m.\n",
      "v\n",
      "wrong.\n",
      "continent.\n",
      "j\n",
      "fire.\n",
      "Cradock.\n",
      "beer-garde-\n",
      "automobile.\n",
      "provid\n",
      "tired.\n",
      "readmittei\n",
      "applause.\n",
      "removed.\n",
      "owners.\n",
      "tl\n",
      "their\n",
      "have.\n",
      "Pleas.\n",
      "tions.\n",
      "is\n",
      "while,\n",
      "Dawson.\"\n",
      "Marne'.\n",
      "call.\n",
      "weeks.\n",
      "M.\n",
      "at\n",
      "-\n",
      "respect.\n",
      "court.\"\n",
      "Gazette.\n",
      "wall.\n",
      "water.\n",
      "i,\n",
      "1902.\n",
      "employed.\n",
      "excitement.\n",
      "declined.\n",
      "paid.\n",
      "freedom.\n",
      "toguardthr\n",
      "as\n",
      "position.\n",
      "boxes.\n",
      ".struggle.\n",
      "Constitution.\n",
      "lUalUc.\n",
      "as\n",
      "breakfast.\n",
      "r-\n",
      "lows:\n",
      "slave-trade-\n",
      "tenni.\n",
      "to\n",
      "more.\n",
      ".\n",
      "deserved.\n",
      "reference.\n",
      "world.\n",
      "States\n",
      "stantia-\n",
      "expedient.\n",
      "welcome.\n",
      "bright\n",
      "and\n",
      "fice.\n",
      "s\n",
      "toe,\n",
      "started.\n",
      "Cuba.\n",
      "taken.\n",
      "Union.\n",
      "that.\"\n",
      "Nicaragua.\n",
      "dition.\n",
      "Newark.\n",
      "andItoldhimsoandhesaidhe\n",
      "d-\n",
      "Ac.,\n",
      "obligation.\n",
      "Rivers.\n",
      "happened.\n",
      "possible.\n",
      "loyally.\n",
      "Union.\n",
      "vessel.\"\n",
      "me.\n",
      "saries.\n",
      "tliu\n",
      "-\n",
      "n\n",
      "at\n",
      "people.\n",
      "brother,\n",
      "civilization.\n",
      "would\n",
      "distress.\n",
      "simplicity.\n",
      "gentle\n",
      "week.\n",
      "state.\"\n",
      "exactly.\n",
      "Island\n",
      "he\n",
      "Ward.\n",
      "our\n",
      "-\n",
      "cents!\n",
      "cused.\n",
      "s\n",
      "s,\n",
      "security.\n",
      "school.\n",
      "broad\n",
      "to\n",
      "Nov\n",
      "Delaware,\n",
      "Bonds.\n",
      "o.\n",
      "reality.\n",
      "sound;\n",
      "written.\n",
      "5005peg\n",
      "himself.\n",
      "peace,\n",
      "bereavement.\n",
      "iiresular,\n",
      "be\n",
      "promises.\n",
      "home.\n",
      "over.\n",
      "itself.\n",
      "I).,\n",
      "unfortunate\n",
      "gogue.\n",
      "the\n",
      "contract\n",
      "lrienu,\n",
      "reported.\n",
      "price.\n",
      "enterprise.\n",
      "of\n",
      "game.\n",
      "baud*,\n",
      "American\n",
      "this\n",
      "want.\n",
      "Jcpson\n",
      "attempts.\n",
      "however.\n",
      "problem.\n",
      "ilianki.\"\n",
      "73\n",
      "ago.\n",
      "noon.\n",
      "with\n",
      "privilege.\n",
      "railroads.\n",
      "a\n",
      "!\n",
      "law.\n",
      "muon.\n",
      "place.\n",
      "season.\n",
      "grind.\n",
      "ever\n",
      "ground\n",
      "Esq.\n",
      "a\n",
      "California.\n",
      "of.\n",
      "public.\n",
      "weather.\n",
      "vestigation.\n",
      "summary:\n",
      "feet.\"\n",
      "heads.\n",
      "'\n",
      "nhaae.\n",
      "ruWKiM\n",
      "■,,'\n",
      "for\n",
      "by.\n",
      "Calvin\n",
      "charge.\n",
      "mountains.\n",
      "arc\n",
      "of\n",
      "larger,\n",
      "debater.”\n",
      "war.\n",
      "said:\n",
      "1370,\n",
      "Chesapeake.\n",
      ".\n",
      "fort.\n",
      "by\n",
      "e\n",
      "Is.\n",
      "1903.\n",
      "high.\n",
      "UK,\n",
      "majority.\n",
      "and\n",
      "result.\n",
      "er\n",
      "?\"\n",
      "Lancaster,\n",
      "slam.\"\n",
      "(Signed)\n",
      "Weekly.\n",
      "'\n",
      "tables.\"\n",
      "Moines.\n",
      "ObBorver,\n",
      "one.\n",
      "thought.\n",
      "forfeited.\n",
      "to\n",
      "It\n",
      "is\n",
      "Chronicle.\n",
      "it.\n",
      "World.\n",
      "d\n",
      "way\n",
      "night.\n",
      "\"Congress.\"\n",
      "hkeecape.\n",
      "Calhoun',\n",
      ".\n",
      "should\n",
      "the\n",
      "and\n",
      "1\n",
      "elsewhere.\n",
      "ion.\n",
      "vonom\n",
      "climate.\n",
      "both.\n",
      "Lawrence.\n",
      "iionne-\n",
      "co!\n",
      "expedition.\n",
      "contains.\n",
      "and\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Burgess)\n",
      "cheap.\n",
      "ENOUGH\n",
      ",\n",
      "lor\n",
      "by\n",
      "over.\n",
      "JAA41»\n",
      "tory\n",
      "belonging,\n",
      "on\n",
      "style:\n",
      "L'XJK'IulitUrfH.\n",
      "bill!\n",
      "subscription\n",
      "in-\n",
      "crats.\n",
      "them.\n",
      "tho\n",
      "velt\n",
      "furnished.\n",
      "acts.\n",
      "ers.\n",
      "wasted\n",
      "Nevada,\n",
      "harbor.\n",
      "surrendered\n",
      "see\n",
      "knives.\n",
      "evei\n",
      "world:\n",
      ".\n",
      "Promotions.\n",
      "\"\n",
      "ways.\n",
      "buried\n",
      "extra.\n",
      "J.\n",
      "promote.\n",
      "nay.\n",
      "dr.\n",
      "less.\n",
      "'coop.'\n",
      "Country.\n",
      "wife.\n",
      "rot.\"\n",
      "calculated.\n",
      "education.\n",
      "person,\n",
      "»\n",
      "children.\n",
      "ILLY,\n",
      "Bear.\n",
      "ways.\n",
      "!«-\n",
      "and\n",
      "management.\n",
      "ants.\n",
      "commodity.\n",
      "scenes\n",
      "tribune.\n",
      "wide-sprea-\n",
      "costs.\n",
      "success.\"\n",
      "ter\n",
      "indopen-deuce-\n",
      "a\n",
      "La'jle.\n",
      "after\n",
      "pose.\n",
      "lauding.\n",
      "llowers,\n",
      "Hoursfrom9a.m..todr.m.\n",
      "year.\n",
      "Keswick's\n",
      "forever\n",
      "us,\n",
      "witness\n",
      "Spetoh\n",
      "living-situation.\n",
      "e\n",
      ".\n",
      "other.\n",
      "helpful.\n",
      "another.\n",
      "exhib.:\n",
      "best\n",
      "so\n",
      "animated.\n",
      "says\n",
      "arrests.\n",
      "abiding\n",
      "America.\n",
      "¬\n",
      "t\n",
      "says:\n",
      "efface.\n",
      "hereafter.\n",
      "fit.”\n",
      "Iwentv-tdne\n",
      "eorners\n",
      "I.\n",
      "the\n",
      "night.\n",
      "immediately.\n",
      "law.\n",
      "club.'\n",
      "editors.\n",
      "McDonough\n",
      "Stomp,\"\n",
      "?\n",
      "as\n",
      "erty.\"\n",
      "evening,\n",
      "manafar-\n",
      "C.\n",
      "distress.\n",
      "(SKAL.I\n",
      "States.\n",
      "without.\n",
      "e\n",
      "school.\n",
      "dealers.\n",
      "gs\n",
      "thing.\n",
      "names.\n",
      "fo\n",
      "opposed.\n",
      "parent\n",
      "greatly\n",
      "ber\n",
      "lce.\n",
      "their\n",
      "privilege.\n",
      "Cost*.\n",
      "Wlikes-Barr-\n",
      "Sentinel.\n",
      "13.\n",
      "Patriot.\n",
      "night.\n",
      "under\n",
      "war.\n",
      "com-\n",
      "tion.\n",
      "y\n",
      ".\n",
      "victim\n",
      "sharks.\n",
      "by\n",
      "day.\n",
      "graved.\n",
      "other\n",
      "one\n",
      "Department.\n",
      "the\n",
      "Trado\n",
      "monuments\n",
      "accident.\n",
      "port.\n",
      "anticipated.\n",
      "through\n",
      "society.\n",
      "soon.\n",
      "llechuanas.\n",
      "holidays\n",
      "rr\n",
      "insisted\n",
      "snry,\n",
      "Tribune.\n",
      "can\n",
      "father-in-la-\n",
      "hoards.\n",
      "Assembly.\n",
      "e\n",
      "fortunate.\n",
      "East.\n",
      "Gregory,\n",
      "floners\n",
      "more\n",
      "These\n",
      "stove.\n",
      "overabundance\n",
      "a\n",
      "your-icou-\n",
      "period.\n",
      "ago.\n",
      "them.\n",
      "tug.\n",
      "them.\n",
      "work.\n",
      "tbo\n",
      "s\n",
      "here.\n",
      "semi-annuall-\n",
      "tion.\n",
      "desirable.\n",
      "country.\n",
      "board\n",
      "cincts.\n",
      "President.\n",
      "lf\n",
      "period.\n",
      "pit.\n",
      "sleep.\n",
      "here.\"\n",
      "do,\n",
      "a\n",
      "cousin\n",
      "lookout.\n",
      "days.\n",
      "wages?\"'\n",
      "DosAngcs,\n",
      "t\n",
      "female,\n",
      "him.\n",
      "Secretary\n",
      "Sampson.\n",
      "A\n",
      "time.\n",
      "President?\n",
      "lost.”\n",
      "mistross.\n",
      "vote.\n",
      "tearful.\n",
      "Juneau.\n",
      "you?\n",
      "concerned.\n",
      "perseverence.\n",
      "a\n",
      "Amrrian.\n",
      "flattery.\"\n",
      "market.\n",
      "love.\n",
      "years.\n",
      "battle:\n",
      "oi\n",
      "without\n",
      "-\n",
      "papers\n",
      "wealthy.\n",
      "Newt.\n",
      "tin.\n",
      "Interest\n",
      "fashion\n",
      "togetner,\n",
      "lution^\n",
      "shore.\n",
      "re-\n",
      ".\n",
      "lectators.\n",
      "death.\n",
      "such\n",
      "betteer.\n",
      "control.\n",
      "thereof.\n",
      "standing\n",
      "nt\n",
      "crystal.\n",
      "payroll.\n",
      "opinion.\n",
      "is\n",
      "’\n",
      "membership.\n",
      "Keehn.\n",
      "premises.\n",
      "people,\n",
      "suspended\n",
      "name.\n",
      "deserted,Ibrlio\n",
      "consumption.\n",
      "not\n",
      "accruing\n",
      "Hepnblio,\n",
      "iury.\n",
      "as\n",
      "shrine.\n",
      "H\n",
      "world/'\n",
      "revoked,\n",
      "points.\n",
      "hand.\n",
      "Porcupine\n",
      "ago.\n",
      "reasonable.\n",
      "were\n",
      "geese.\n",
      "there,\n",
      "geography.\n",
      "fering.\n",
      "etc.,\n",
      "-w-\n",
      "\"cause.\"\n",
      "ble\n",
      "in\n",
      "Observer.\n",
      "15,1888.\n",
      "fire.\n",
      "standpoint.\n",
      "and\n",
      "iut-.\n",
      "accident\n",
      "contented.\n",
      "she\n",
      "s\n",
      "Dukes.\n",
      "fur.\n",
      "s,\n",
      "rifle.\n",
      "some\n",
      "panies\n",
      "drive.\n",
      "i\n",
      "roads.\n",
      "the\n",
      "it.\n",
      "Science.\n",
      "lion.\n",
      "f,\n",
      "to\n",
      "be-\n",
      "ty\n",
      "u-\n",
      "In.\"\n",
      "-\n",
      "to\n",
      "reach\n",
      "provided-\n",
      "convention.\n",
      "line.\n",
      "4th\n",
      "achieve¬\n",
      "Gazette.\n",
      "*50.000.\n",
      "methods.\n",
      "nodded.\n",
      "two\n",
      "vassals\n",
      "origin.\n",
      "sword.\n",
      "Loos.\n",
      "Paris\n",
      "6\n",
      "Record.\n",
      "'2o.\n",
      "<\n",
      "ear.\"\n",
      "stated.\n",
      "people\".\n",
      "investigation,\n",
      "attraction.\n",
      "highways.\n",
      "court.\n",
      "senseless!\n",
      "Magazine.\n",
      "ashes.\n",
      "once\n",
      "quences.\n",
      "corners.\n",
      "are\n",
      "city.\n",
      "school\n",
      "dis-\n",
      "leader.\n",
      "Slates\n",
      "spring.\n",
      "potior.\n",
      "the\n",
      "beginning.\n",
      "ford*.\n",
      "14.\n",
      "us.\n",
      "disgust.\n",
      "fork\n",
      "launched.\n",
      "cc\n",
      "market.\n",
      ".\n",
      "fee\n",
      "etc.\n",
      "now.\"\n",
      "rate\n",
      "it.\n",
      "4.\n",
      "oliM-rvi-\n",
      "notwithstanding.\"\n",
      "High.\n",
      "back.\n",
      "habit.\n",
      "taller.\n",
      "candidate.\n",
      "doctors.\"\n",
      "just.\n",
      "work.\n",
      "hi-\n",
      ".\n",
      "We\n",
      "fair\n",
      "cards.\n",
      "banks\n",
      "X.\n",
      "terrible.\n",
      "-\n",
      "tlon.\n",
      "ago,\n",
      "city.\n",
      "water.\n",
      "when\n",
      "anybody.\n",
      "cease.\n",
      "telescopes\n",
      "said\n",
      "circumstance,\n",
      "coun­\n",
      "s\n",
      "Iceland\n",
      "advancing.\n",
      "Prineviile.\n",
      "remedy.\n",
      "say.\n",
      "on-\n",
      "Mr.\n",
      "carried.\n",
      "paid,\n",
      "future.\"\n",
      "graced.\n",
      "truth.\n",
      "yeat--\n",
      "?”\n",
      "preserves\n",
      "such\n",
      "home.\n",
      "d\n",
      "waa\n",
      "phy.\"\n",
      "Times:\n",
      "in-\n",
      "footprints\n",
      "funeral.\n",
      "present.\n",
      "crossing\n",
      "robber,\n",
      "diamonds\n",
      "sort.\n",
      "friends\n",
      ":\n",
      "Times.\n",
      "Post.\n",
      "societies.\n",
      "hunger,\n",
      "avarice.\n",
      "it.\n",
      "right*\n",
      "e\n",
      "taken.\n",
      "\"liome-tralner-\n",
      "again.\n",
      "conception.\n",
      "good.\n",
      "Science.\n",
      "feeling.\n",
      "remarked,\n",
      "Smith.\n",
      "present.\n",
      "ingredients.\n",
      "pre-\n",
      "ju\n",
      "sold:\n",
      "580pm.\n",
      "ten-\n",
      "somo-\n",
      "lf\n",
      "shall\n",
      "Mexican\n",
      "life.\n",
      "cure.\n",
      "married\n",
      "high-cla-\n",
      "day.\n",
      "a\n",
      ",\n",
      "me.\n",
      "lino\n",
      "spot.\n",
      "at-\n",
      "by\n",
      "aeason.\n",
      "accepted.\n",
      "justice.\n",
      "The\n",
      "nations.\n",
      "away.\n",
      "to.\n",
      "Cal.,\n",
      "country.\n",
      "confessed.,\n",
      "issue.\n",
      "end.\n",
      "al\n",
      "Christ.\n",
      "Auguit\n",
      "nations.\n",
      "c\n",
      "trackage.\n",
      "mine-\n",
      "desperado.\n",
      "won.\n",
      "thereby.\n",
      "ravenous.\n",
      "shot.\n",
      "newspapers\n",
      "b62.\n",
      "suf-\n",
      "are\n",
      "friends\n",
      "defrauded.\n",
      "contracted.\n",
      "lesson.\n",
      "remunerative\n",
      "Kempls:\n",
      "institutions.\n",
      "icate.\n",
      "y\n",
      "pleasant.'\n",
      "continuously.\n",
      "tell.\n",
      "goest.\n",
      "producer.\n",
      "thank\n",
      "•\n",
      "States.\n",
      "cured.\"\n",
      "Navy\n",
      "M.\n",
      "asser\n",
      "Judgment\n",
      "course.\n",
      "1862.\n",
      "township\n",
      "case.\n",
      ".\n",
      "y\n",
      "a\n",
      "made.\n",
      "made.\n",
      "n\n",
      "acres,\"\n",
      "them.\n",
      "wonderlngs.\n",
      "mind\n",
      "collts-tlv-\n",
      "ter.\n",
      "nature.\n",
      "tion.\n",
      "improved.\n",
      "the\n",
      "\"\n",
      "to-w-\n",
      "Paper,\n",
      "broidery.\n",
      "the\n",
      "X'llt.\n",
      "an\n",
      "often.\n",
      "that\n",
      "of\n",
      "cemetery.\n",
      "depression\n",
      "\"\n",
      "it.\n",
      "hotel.\n",
      "IS\n",
      "a\n",
      "m.\n",
      "render.\n",
      "back-\n",
      "themselves.\n",
      "c-\n",
      "a\n",
      "*ore’\n",
      "companies.\n",
      "the\n",
      "”\n",
      "manner.\n",
      "tha\n",
      "Granger.\n",
      "difficol'7>\n",
      "Lake\n",
      "placed\n",
      "lf\n",
      "for\n",
      "lllll\n",
      ".\n",
      "learn.\n",
      "so.\n",
      "the\n",
      "-\n",
      "again.\n",
      "sec-\n",
      "tlmo.\n",
      "isfied.”\n",
      "'\n",
      "GrIL\n",
      "government.\n",
      "democrats,\n",
      "pit.\n",
      "even.\n",
      "McCausland.\n",
      "consideration.\n",
      "fragments.\n",
      "the-\n",
      "Colombia\n",
      "defendant.\n",
      "like\n",
      "outrage-\n",
      "yours.\n",
      "Adjourned.\n",
      "1\n",
      "church.\"\n",
      ",\n",
      "People.\n",
      "STORY\n",
      "revolver\n",
      "contract\n",
      "-\n",
      "nttii-\n",
      "Adjourned.\n",
      "him.\"\n",
      "dee.\"\n",
      "mous.\n",
      "bars\n",
      "24:10.\n",
      "inde-\n",
      "road.\"\n",
      "is.\n",
      "a.\n",
      "-\n",
      "remedy.\n",
      "in\n",
      "s\n",
      "match.”\n",
      "Laughter.\n",
      "that\n",
      "aforeiaii’,\n",
      "accident.\n",
      "—Exchange.\n",
      "and\n",
      "whiter.\n",
      "frc--\n",
      "week.\n",
      "party.\n",
      "Collins\n",
      "st\n",
      "•\n",
      ".\n",
      "departure.\n",
      ";\n",
      "stock-raisin-\n",
      "of\n",
      "large.\n",
      "completed\n",
      "wrong.\n",
      "mines.\n",
      "outside.\n",
      "lower\n",
      "avoided.\n",
      "me\n",
      "means.\n",
      "Hcdjuz,\n",
      "tier?\n",
      "towns—wht'je\n",
      "quality.\n",
      "cure.\n",
      "which,\n",
      "ever.\n",
      "of\n",
      "shall\n",
      "\"\n",
      "—Am,\n",
      "bellovo\n",
      "14.\n",
      "im-\n",
      "expenses.\n",
      "called\n",
      "43,4■\n",
      "Herald.\n",
      "rain,\n",
      "mo-\n",
      "money-mak-\n",
      "celebration.\n",
      "es\n",
      "-\n",
      "miles.\n",
      "friends.\n",
      "m.\n",
      "their\n",
      "dustry,\n",
      "opposition.-\n",
      "damage.\n",
      "atten-\n",
      "ance\n",
      "bargain,?\n",
      "17.\n",
      "yourself.\n",
      "no\n",
      "M\n",
      "allirmod.\n",
      "yours,\n",
      "the-\n",
      "enjoyed.\n",
      "Taj.\n",
      "largely.\"\n",
      "d\n",
      "pl-\n",
      "vore.\n",
      "march.\n",
      "houses.\n",
      "onth■\n",
      "nfitasfwl\n",
      "1876.\n",
      "44.\n",
      "tense\n",
      "times.\n",
      "or\n",
      "1878.\n",
      "23\n",
      "accommodations.\n",
      "Tops-ha-\n",
      "such\n",
      "hats.\n",
      "doc-\n",
      "plant.\"\n",
      "e\n",
      "he\n",
      "truth.\n",
      "him\n",
      "PUUH,\n",
      "SOUTH\n",
      "it.\"\n",
      "-ce\n",
      "at\n",
      "are\n",
      "granjes.\n",
      "co-couuaei.\n",
      "former.\n",
      "of\n",
      "it\n",
      "prayed.\n",
      "mentioned,\n",
      "them.\n",
      "1890.\n",
      "m,\n",
      "ery.\n",
      "full.\n",
      "votes.\n",
      ">\n",
      "flna\n",
      "Inent.\n",
      "revc-\n",
      "75\n",
      "lefL\n",
      "repealed.\n",
      "toQO.\n",
      "camps.\n",
      "herder.\n",
      "Centre.\n",
      "error.\n",
      "mommy\n",
      "School.\n",
      "Week.\n",
      "sea.\n",
      "night.\n",
      "troops.\n",
      "work,and\n",
      "serv't\n",
      "production.\"\n",
      "clBe.\n",
      "majority.\n",
      "holidays.\n",
      "year.\n",
      "d\n",
      "cients.\n",
      "&c.,\n",
      "1893.\"\n",
      "chargo\n",
      "cattle.\"\n",
      "Now\n",
      ".\n",
      "on.\"\n",
      "executive\n",
      "goods.\n",
      "Asam.\n",
      "property.\n",
      "p\n",
      "have\n",
      "analgin*.\n",
      "the\n",
      "kc.\n",
      "$700,000,000.\n",
      "taklrg\n",
      "it,\n",
      "wan\n",
      "ioint.\n",
      "cause.\n",
      ".!\n",
      "f<\n",
      "h\n",
      "quantities.\n",
      "services\n",
      "when.\n",
      "*t\n",
      "tea,\n",
      "ments.\n",
      "ngovern-\n",
      "state.\n",
      "Post.\n",
      "city.\n",
      "In-\n",
      "charges.\"\n",
      "trip.\n",
      "2.\n",
      "a\n",
      "tho\n",
      "yet.\n",
      "race.\n",
      ",\n",
      "postage.\n",
      "thanksgiving\n",
      "that;,\n",
      "was\n",
      "190D.\n",
      "York.\n",
      "abdomen,\n",
      "conventions\n",
      "back.\n",
      "Affiux.\n",
      "baken,\n",
      ":\n",
      "sites.\n",
      "wa-\n",
      "io-\n",
      "msgizines.\n",
      "time.\n",
      "alone,\n",
      "craiy.\"\n",
      "drugs.\n",
      "liking.\n",
      "season.\"\n",
      "tower.\n",
      ".\n",
      "silenced.\n",
      "from.\n",
      "court;\n",
      "Alaska.\"\n",
      "effective.\n",
      "neck.\n",
      "not\n",
      "paid.\n",
      "short\n",
      "Esq.\n",
      "herein.\n",
      "thorti.\n",
      "l\n",
      "void.\n",
      "over.\n",
      "difficulty.\n",
      "it.\n",
      "cities.\n",
      "Investigate\n",
      "confusion.\n",
      "insurance\n",
      "otm.\n",
      "t\n",
      "since.\n",
      ".\n",
      "away.\n",
      "bers.\n",
      "Harding,\n",
      "fear.”\n",
      "persevering,\n",
      "Philippines.\n",
      "alley,\n",
      "house.\n",
      "ores.\n",
      ".\n",
      "easier\n",
      "Wot\n",
      "him.\n",
      "treatment\n",
      "tbe\n",
      "flub;\n",
      "patriotism.\n",
      "cheap.\n",
      "boa\n",
      "....\n",
      "for\n",
      "cordingly.\n",
      "1\n",
      "lAdtfree..\n",
      "land.\n",
      "a,\n",
      "glory.\n",
      "ao\n",
      "aloe\n",
      "grain.\n",
      "him.\n",
      "demanded.\n",
      "John\n",
      "cash.\n",
      ".\n",
      "Bussians.\n",
      "expenditures.\n",
      "forenoon.\n",
      "weak.\n",
      "sarcastically.\n",
      "state.\n",
      "of\n",
      "N.Y.\n",
      "Ms.\n",
      "s\n",
      "on\n",
      "him\n",
      "lake.\n",
      "per-\n",
      "cial.\n",
      "Lorraine.\n",
      "th\n",
      "husband\n",
      "\"Deaconesses.\"\n",
      "—I'hicai/o\n",
      "$75.\n",
      "o\n",
      "it.\n",
      "!\n",
      "them.\n",
      "satisfied.\n",
      "rifle.\n",
      "trfct.\n",
      "liberal\n",
      "rules!\n",
      "printer.\n",
      "spring.\n",
      "four\n",
      "are\n",
      "that\n",
      "fair.\n",
      "governor-elect.\n",
      "advantages.\n",
      "specifications\n",
      "localities.\n",
      "midnight.\n",
      "explanatory\n",
      "1852.\n",
      "ino.\n",
      "presents.\n",
      "wharfage.\n",
      "rs\n",
      "cu\n",
      "of\n",
      "15\n",
      "paper.\n",
      "it.\n",
      "experience.\n",
      "vacation.\n",
      "are.\"\n",
      "goods.\n",
      "home\n",
      "bear-tra-\n",
      "themselves,\n",
      "besides.\n",
      "Labor..\n",
      "I\n",
      "hits.\n",
      "admitted\n",
      "full\n",
      "trT\"\"\n",
      "to\n",
      ">\n",
      "rough.\"\n",
      "verse\n",
      "approve.\n",
      "intheh11.\n",
      "Pickett.\n",
      "on\n",
      "doubtful\n",
      "revive^ller,\n",
      "pan\n",
      "busi¬\n",
      "remarks:\n",
      "fail\n",
      "hand.\n",
      "thcui.\n",
      "ago.\n",
      "g\n",
      "defeated.\n",
      "Wo\n",
      "-\n",
      "investigating\n",
      "pounds.\n",
      "!\n",
      "commissioners.\n",
      "inten-\n",
      "Rankin\n",
      "single-dro-\n",
      "Shorten.\n",
      "five\n",
      "seven\n",
      "schools.\n",
      "stored.\n",
      "conveyed.\n",
      "to\n",
      "mission\n",
      "up.\"\n",
      "timbers.\n",
      "remedy.\n",
      "a\n",
      "copy.\n",
      "followsi\n",
      "Home.\n",
      "beginning.\n",
      "\"Darling,\n",
      "Comuion-\n",
      "above.\n",
      "Taxes\n",
      "gloom\n",
      "In.\n",
      "writing.\"\n",
      "day.\n",
      "corrected.\"\n",
      "Gregg,\n",
      "tho\n",
      "vested.\n",
      "him.\n",
      "havo\n",
      "again.\n",
      "$1,155,000.\n",
      "Newark.\n",
      "profess.\n",
      "follows:\n",
      "having.\n",
      "members?\n",
      "of\n",
      "It\n",
      "Men.\"\n",
      "copy.\n",
      "truly,\n",
      "integrity.\n",
      "system.\n",
      "follows:\n",
      "now\n",
      "army.\n",
      ")\n",
      "went.\n",
      "expected.\n",
      "2\n",
      "center.\n",
      "llrvi<v\\\\\n",
      "Express.\n",
      "Esq.\n",
      "church.\n",
      "power.\n",
      "abuses.\n",
      "18/.'I.\n",
      "people.\n",
      "c.il\n",
      "locality.\n",
      "estate.\n",
      "\"\n",
      "ness.\n",
      "\"book.\n",
      "price.\n",
      "crowd.\n",
      "a\n",
      "had.\n",
      "landlord.\"\n",
      "a\n",
      "tlj!(\n",
      "control.\n",
      "Gray-\n",
      "pie.\n",
      "mine.\n",
      "edition,\n",
      "»—■\n",
      "in\n",
      "us.\n",
      "crime.\n",
      "personal\n",
      "the\n",
      "dollars.\"\n",
      "following:\n",
      "Their\n",
      "hist\n",
      "pioneering.\n",
      "him\n",
      "r\n",
      "no-\n",
      "s\n",
      ".\n",
      "30\n",
      "Times.\n",
      "State\n",
      ",,\n",
      "ordinance.\n",
      "a\n",
      "welfare.\n",
      "o\n",
      "Aat..\n",
      "world.\"\n",
      "impelled\n",
      "Germany.\n",
      "s\n",
      "case.\n",
      "the\n",
      "scat-\n",
      "Commission.-\n",
      "15.\n",
      "along.\"\n",
      "loss.\n",
      "mothers.\n",
      "seemed\n",
      "finances.\n",
      "fail.\n",
      "shape.\n",
      "It\n",
      "it.\n",
      "Ted.\n",
      "Uuildim.*'.\n",
      "Pa.\n",
      "others.\n",
      "twenty-\n",
      "Ward.\n",
      "Guyton.\n",
      "reward.\n",
      "denominations\n",
      "schedule.\n",
      "to\n",
      "who\n",
      "looked\n",
      "her:\n",
      "jt\n",
      "-\n",
      "t\n",
      "memory\n",
      "dismissed.\n",
      "lie.\n",
      "\"\n",
      "heads.\n",
      "drsas\n",
      "firm.\n",
      "feelings.\n",
      "vnpir\n",
      "address;\n",
      "fering.\n",
      "forfeited.\n",
      "purose.\n",
      "summer.\n",
      "aforesaid.\n",
      "given.\n",
      "ground\n",
      "him.\n",
      "bu-\n",
      "relatives\n",
      "tion.\n",
      "instruments,Btruddaa.\n",
      "York.\n",
      "high.\n",
      "yours,\n",
      "I\n",
      "denial.\n",
      "t-~»\n",
      "communicable.\n",
      "pe*s,\n",
      "er.\n",
      "complaint.\n",
      "Judgment.\n",
      "d\n",
      "Lode.\n",
      "installed.\n",
      "so.\n",
      "y\n",
      "be.\n",
      "fou\n",
      "strip.\n",
      "2inthiOi\n",
      "Division,\"\n",
      "to\n",
      "objections.\n",
      "”\n",
      "pealed.\n",
      "of\n",
      "assist\n",
      "two.\n",
      "-\n",
      "h\n",
      "brium.\n",
      "Willis.\n",
      "more.\n",
      "etc.,\n",
      "ages.'\n",
      "good.”\n",
      "the\n",
      "Positive\n",
      "witness.\"\n",
      "therefor.\n",
      "it\n",
      "explo-\n",
      ",\n",
      "u\n",
      "robber*.\n",
      "deed,\n",
      "e\n",
      ".\n",
      "Digger.\n",
      "—Balti-\n",
      "Adjourned.\n",
      "$22,-\n",
      "liiinlc\n",
      "land\n",
      "about\n",
      "epee.\n",
      "days.\"\n",
      "in.\n",
      "murders.\n",
      "confidence.\n",
      "approved.\n",
      "played.\n",
      "away.\n",
      "festival.\n",
      "evil.\n",
      "but-\n",
      "woo-\n",
      ".\n",
      "tiroes,\n",
      "offering\n",
      "avenue.\n",
      "legs.\n",
      "sleep\n",
      "health.\n",
      "suehftft,\n",
      "Davidson.\n",
      "work,\n",
      "»\n",
      "it.\"\n",
      "forth.\n",
      "proportionate\n",
      "Italy\n",
      "it.\n",
      "force.\n",
      "In\n",
      "more.\n",
      "with\n",
      "oh-\n",
      "great.\n",
      "gentleman.\n",
      "fright.\n",
      "harem.\n",
      "fallen.\n",
      "ladies.\n",
      "days.\n",
      "management?\n",
      "adjourned.\n",
      "duties.\n",
      "military.\n",
      "llv\n",
      "ight\n",
      "In\n",
      "paid\n",
      "mate.\n",
      "prince?'\n",
      "apology.\n",
      "Democrat.\n",
      "love.\n",
      "SIOO.\n",
      "of\n",
      "effect.\n",
      "experience.\n",
      "E.\n",
      "coupon\n",
      "Art\n",
      "land.\n",
      "and\n",
      "America.\n",
      "Register.\n",
      "g\n",
      "nent.\n",
      "community.\n",
      "aadertak-\n",
      "lino\n",
      "lishment.\n",
      "party.\n",
      "not\n",
      "question.\n",
      "coat-front.\n",
      "i\n",
      "st\n",
      "to.\n",
      "fail.\n",
      "conveniences\n",
      "occurs,\n",
      "buildings.\n",
      "e\n",
      "sufferers.\n",
      "filing.\n",
      "inortgag-e-\n",
      "adjourned.\n",
      "Schindeldecker,\n",
      "govern-\n",
      "Slates.\n",
      "and\n",
      "discarded.\n",
      "snow.\n",
      "righteousness\n",
      "toui.\"\n",
      "Government.\"\n",
      "the\n",
      "others,\n",
      "Scribner's.\n",
      "Senator,\n",
      "assisted.\n",
      "ikoro.\n",
      "custody.\n",
      "ing\n",
      "exercises\n",
      "Paris.\n",
      "answer.\n",
      "him.\"\n",
      "s\n",
      "War.\n",
      "purpose.\n",
      "State.\n",
      "Comment.\n",
      "health.\n",
      "follies.\n",
      "Mrs.\n",
      "results.\n",
      "-\n",
      ",\n",
      "to\n",
      "follow,\n",
      "tlic\n",
      ".*---V\n",
      "said.\n",
      "wicked-\n",
      "following\n",
      "to\n",
      "iniform.\n",
      "ly\n",
      "theory.\n",
      "theory.\n",
      "three-fourth-\n",
      "water.\n",
      "Jr.,\n",
      "it.\n",
      "Valley.\n",
      "Thee.\"\n",
      "communicable.\n",
      "health.\n",
      "date\n",
      "nn\n",
      "sions.\n",
      "matter.\n",
      "perquisites.\n",
      "tucky\n",
      "«fec.,\n",
      "ing.\n",
      "them.\n",
      "business.\n",
      "down.\n",
      "each.\n",
      "-\n",
      "inferior\n",
      "home.\n",
      "4t—36\n",
      "thereat.\n",
      "newest.\n",
      "t\n",
      "Navy.\n",
      "filled.\n",
      "and\n",
      "land.\"\n",
      "delay.\n",
      "load.\"\n",
      "family.\n",
      "Uaicllt,\n",
      "118\n",
      ".\n",
      "ones.\n",
      "relatives.\n",
      "work,\n",
      "same.\n",
      "gathered.\n",
      "American.\n",
      "died?\"\n",
      "Honor.\n",
      "A.\n",
      "tho\n",
      "population,\n",
      "years.\n",
      "eternal\n",
      "maoleod\n",
      "court.\n",
      "kind.”\n",
      "lefthim.\n",
      "st.,\n",
      "valor.\n",
      "shippers.\n",
      "wire\n",
      "the\n",
      "day.\n",
      "Xvavs.\n",
      "ers.\n",
      "equipment\n",
      "commendable.\n",
      "Huntsfcille:\n",
      "man.\"\n",
      "Guild.\n",
      "ment\n",
      "printer.\n",
      "remedy.\n",
      "souvenir\n",
      "here\n",
      "Seat-\n",
      "stock.\n",
      "msue\n",
      "appearance.\n",
      "ooo-\n",
      "the\n",
      "vestigation.\n",
      "wait.\n",
      "privileges.\n",
      "a\n",
      "\"stuck\"\n",
      "Ameri-\n",
      "circles.\n",
      "seded.\n",
      "text\n",
      "Bill\n",
      "relief\n",
      "exporti\n",
      "no1\n",
      "ing.\n",
      "could\n",
      "streets\n",
      "Bank-wher-\n",
      "*'\n",
      "destructive\n",
      "pound.\n",
      "world.\n",
      "notv\n",
      "preserves.\n",
      "School.\n",
      "smile.\n",
      "delay.\n",
      "King.\"\n",
      "wed.\n",
      "res--\n",
      "Sugarloaf.\n",
      "Baltimore.\n",
      "relief.\n",
      "contracted-\n",
      "information.\n",
      "tobacco.\n",
      "past.\n",
      "The\n",
      "Company\n",
      "the\n",
      "head\n",
      "parlor.\n",
      "assistance.\n",
      "inator.\n",
      "And\n",
      "couple.\n",
      "!\n",
      "time.\n",
      "it.\n",
      "re-\n",
      "slavery,.\n",
      "powers.\n",
      "kennel.\n",
      "Clerk\n",
      ",\n",
      "him.\n",
      "yellow.\n",
      "cured.\"\n",
      "Chester.\n",
      "eliminated\n",
      "and\n",
      "success.\n",
      "*\n",
      "center.\"\n",
      "!\n",
      "him.\n",
      "Gazette.\n",
      "antagonists.\n",
      "law.\n",
      "UDion.\"\n",
      "ble.\n",
      "’\n",
      "Iar?e.\n",
      "estant\n",
      "best.,\n",
      "--\n",
      "ruin\n",
      "will\n",
      "the\n",
      "ment*.\n",
      "measure.\"\n",
      "Aboliuosisa^\n",
      "bones.\n",
      "firtther.\"\n",
      "American.\n",
      "it\n",
      "it.\"\n",
      "circus\n",
      "thanlrRo-Itrlnf-\n",
      "but\n",
      "Fisherman.\n",
      "9\n",
      "containing\n",
      "Ibom.\n",
      "sal-\n",
      "Bnff.ilo\n",
      "News\n",
      "investigation.\"\n",
      "retract.\n",
      "affairs.\n",
      "up-\n",
      "events.\n",
      "settled\n",
      "aggrandizement.\n",
      "iug.\"\n",
      "expenditures.\n",
      "up\n",
      "\"Yes.\"\n",
      "had\n",
      "s,\n",
      "remedy.\n",
      "surpassod.\n",
      "easy\n",
      "herself.\n",
      "-rtrick.\"\n",
      ",\n",
      "mc\n",
      "Cemetery.\n",
      "Weston.\n",
      "colter-\n",
      "fol-\n",
      "sea.\n",
      "good-night!\"\n",
      "potatoes,\n",
      "power\n",
      "*\n",
      "stand\n",
      "if\n",
      "laws.\n",
      "follows:\n",
      "Herbert........\n",
      "mentioned.\n",
      "us.\n",
      "N.\n",
      "winter.\n",
      "the\n",
      "a\n",
      "Tennessee.\n",
      "by\n",
      "being\n",
      "worse\n",
      "machinery.\"\n",
      "affirmed.\n",
      "on--\n",
      "statesmen\n",
      "Creek\n",
      "street\n",
      "co-partnership..\n",
      "'\n",
      "pppreciato.\n",
      "e\n",
      "on\n",
      "tion.\n",
      "d.\n",
      "1887.\n",
      "faithfully,\n",
      "existing.\n",
      "Ing\n",
      "clock.\n",
      "committees.\n",
      "1'.\n",
      "show.\"\n",
      "consistency.\n",
      "rupee.\n",
      "discovered\n",
      "done.\n",
      "Cod.’\n",
      "&Enq.\n",
      "of\n",
      "upon.\n",
      "offer.\n",
      "as\n",
      "evolved.\n",
      "sutreested.\n",
      "himself.\n",
      "known\n",
      "and\n",
      "1\n",
      "children.\n",
      "and\n",
      "his\n",
      "c\n",
      "increased.\n",
      "at\n",
      "e\n",
      "fashionable.\n",
      "months.\n",
      "6o\n",
      "fire.\n",
      "mise\n",
      "it\n",
      "preserves.\n",
      "passage.\n",
      "slavery,\n",
      "BnoxiiKiw.\n",
      "the\n",
      "follows:\n",
      "coin.\n",
      "IT.-:\n",
      "what\n",
      "peace,\"\n",
      "country.\n",
      "Esq.,\n",
      "life.\"\n",
      "innocence.\n",
      "clipper\n",
      "complaint.\n",
      "i\n",
      "man\n",
      "property.\n",
      "art.-\n",
      "-\n",
      "forgets.\"\n",
      "i.i\n",
      "German-Prussia-\n",
      "1H90.\n",
      "a\n",
      "corporations.\n",
      "k.\n",
      "place.\n",
      "West.\n",
      "campaign.\n",
      "congress\n",
      "debt.\"\n",
      "result.\n",
      "-\n",
      "patriots.\n",
      "h\n",
      "town.\n",
      "and\n",
      "the\n",
      "detected.\n",
      "ftatp\n",
      "I\n",
      "syndicates.\n",
      "dignity.\n",
      "it.\n",
      "\"commercial\n",
      "health.\n",
      "question.\n",
      "you,\n",
      "Bird\n",
      "guidance,\n",
      "night.\n",
      "a\n",
      "13,425,000.\n",
      "Connaught.\n",
      "onous.\n",
      "areeu.\n",
      "of\n",
      "st\n",
      "pupils\n",
      "elevens.\n",
      "known.\"\n",
      "tioae\n",
      "tion\n",
      "army;\n",
      "O'Neil.\n",
      "y\n",
      "forever.\n",
      "-\n",
      "Skchvo-\n",
      "task.\n",
      "practiced.\n",
      ".\n",
      "finances.\n",
      "guilty.\n",
      "United\n",
      "suit\n",
      "degree.\"\n",
      "architecture.\n",
      "yours.\"\n",
      "people.\n",
      "bill).\n",
      "killed.\n",
      "besides.\n",
      "spells.\n",
      "from.\n",
      "purchasers.\n",
      "set.\n",
      "(1\n",
      "Inches.\n",
      "remains.\n",
      "-\n",
      "reflected.\n",
      "being,\n",
      "line.\n",
      "-\n",
      "be\n",
      "government\n",
      "Smith.\n",
      "game.\n",
      "you'\n",
      "people.\n",
      "»\n",
      "Wells.\n",
      "Meeker,\n",
      "pleas\n",
      "ticket.\n",
      "thirty-\n",
      "authority\n",
      ",\n",
      "leadership.\n",
      "it\n",
      "im-\n",
      "nincycara.\n",
      "experienced\n",
      "towns.\n",
      "I860.\n",
      "M.\n",
      "midst.\n",
      "reasons.”\n",
      "July,\n",
      "for\n",
      "improvement?\n",
      "ground\n",
      "\"\n",
      "administered.\n",
      "reconstruction.\n",
      "victory.\n",
      "experience.\n",
      "Alaska.\n",
      "artery\n",
      "granted.\n",
      "exclaimed,\n",
      "somevheionn'\n",
      "h,\n",
      "uttered\n",
      "slopes.\n",
      "\"\n",
      "thinking.\n",
      "invader.\n",
      "costs.\n",
      "Lots.\n",
      "in\n",
      "csntidvncs.\n",
      "however,\n",
      "him.\n",
      "full.\n",
      "another.\n",
      "mankind.”\n",
      "blessed.\n",
      "pos-\n",
      "fire.\n",
      "and\n",
      "his-\n",
      "c\n",
      "e\n",
      "to\n",
      "purposes.\n",
      "Treasurer.\n",
      "ge\n",
      "beating.”\n",
      "press.\n",
      "the\n",
      "-\n",
      "oa.\"\n",
      "said:\n",
      "reclaimed.\n",
      "reside.\n",
      "farther.\n",
      ":\n",
      "Trust.\n",
      "count\n",
      "brought\n",
      "dignity.\n",
      "lb.\n",
      "all.\n",
      "it.\n",
      "details.\n",
      "or\n",
      "fallen!\"\n",
      "more.\n",
      "ancestors.\n",
      "tion.\n",
      "\"Twollne.\"\n",
      "ng-\n",
      "empire.\n",
      "o\n",
      "York.\"\n",
      "vention.\n",
      "life-time.\n",
      "molestation.\n",
      "ate.\n",
      "1862.\n",
      "had-submitted.\n",
      "Livingstone.\n",
      "lists.’\n",
      "nt\n",
      "months.\n",
      "Baker's\n",
      "Demacrtt.\n",
      "afternoon.\n",
      "money.\n",
      "discredit.\n",
      "un-\n",
      "official.\n",
      "Douglas.\n",
      "patience.\n",
      "He\n",
      "Utah.\"\n",
      "cent.\n",
      "day*.\n",
      "unanimous.\n",
      "vices.\n",
      "Marvel.\n",
      "wards.\n",
      "as\n",
      "work,\n",
      "1\n",
      "the\n",
      "Baptist.\n",
      "remedy.\n",
      "tight\n",
      "the\n",
      "re\n",
      "America.\n",
      "not.\n",
      "company.\n",
      "own.\n",
      "other.\n",
      "end.\n",
      "the\n",
      "in-law.\n",
      "m.\n",
      "Railroad.\n",
      "sand.\n",
      "in\n",
      "it.\"\n",
      "warranted.\n",
      "coata.\n",
      "congress.\n",
      "money.\n",
      "arrived.\n",
      "by\n",
      "of\n",
      "eight.\n",
      "«\n",
      "men.\n",
      "if\n",
      "I».I.\n",
      "did\n",
      "eager\n",
      "both.\n",
      "that\n",
      "reach.\n",
      "statement.\n",
      "water.\n",
      "measures.\n",
      "leg-\n",
      "of\n",
      "dresses.\n",
      "tone!\n",
      "poopjo.\n",
      "t\n",
      "city,\n",
      "'\n",
      "renew-\n",
      "urofceslonal\n",
      "other.\n",
      "them.\n",
      "American.\n",
      "con-\n",
      "bv\n",
      "have\n",
      ".\n",
      "war.\n",
      "APPROPRIATIONS\n",
      "State.\n",
      "missionary,\n",
      "under.\n",
      "hurt.\n",
      "e\n",
      "vacated.\n",
      "later.\n",
      "uruius,\n",
      "toa\n",
      "■financing.\n",
      "him.'\n",
      "tools.\n",
      "nnemlnn.\n",
      "west.”\n",
      "and\n",
      ">\n",
      "Secretary.\n",
      "tho\n",
      "virtues.\n",
      "1840.\n",
      "God-Saviour.\n",
      "God.\n",
      "eat.\n",
      "Oroville,\n",
      "county.\n",
      "live.\n",
      "party,\n",
      "globe\n",
      "ar*\n",
      "congregation.\n",
      "bearing\n",
      "!».\n",
      "above\n",
      "ideas\n",
      "prayed.\n",
      "Bluffs.\n",
      "investigation.\n",
      "traveler.\n",
      "oil\n",
      "fines.\n",
      "oiteus.\n",
      "disposal.\n",
      "fol-\n",
      ";\n",
      "w&y\n",
      "blood\n",
      "dance\n",
      "protection.\n",
      "Californian.\n",
      "almost\n",
      "them.\n",
      "rot-\n",
      ".Mr.\n",
      "pag.--\n",
      "purchaser.\n",
      "each\n",
      "penalties.\n",
      "dismissed.\n",
      "denoted\n",
      "excuse.\n",
      "o\n",
      "Herald.\n",
      "Gentleman.\n",
      "success.\n",
      "it.\n",
      "methods.\n",
      "factory.\n",
      "St.\n",
      "tion.,\n",
      "an\n",
      "P.\n",
      "cause.\n",
      "Septembers.\n",
      "log.\n",
      "the\n",
      "70^\n",
      "them.\n",
      "ninth.\n",
      "ation.\n",
      "beat.\n",
      "situations.\n",
      "for\n",
      "oi\n",
      "Ghost.\"\n",
      "with\n",
      "Cross.\n",
      "troops.\n",
      "avoided.\n",
      ",\n",
      "sacrifice\n",
      "former.\n",
      "\"Othello\"\n",
      "Philadelphia.\n",
      "power.\n",
      "would\n",
      "gb\n",
      "y\n",
      ".\n",
      "chief.\n",
      "Nicholaieviteh.\n",
      "tho\n",
      "tiling.\n",
      "brnshlng.\n",
      "city.\"\n",
      "consum\n",
      "reform.\n",
      "inteuiioti.\n",
      "bricklaying.\n",
      "forfeited.\n",
      "mate.\n",
      "them.\n",
      "elected\n",
      "Oroville,\n",
      "terms.\n",
      "e\n",
      "order.\n",
      "old\n",
      "favor.\n",
      "them.\n",
      "out.\n",
      "consumption.\n",
      "a.\n",
      "e\n",
      "°r\n",
      "trade.\"\n",
      "$1.\n",
      "just.\n",
      "two\n",
      "charge.\n",
      "adopted.\n",
      "nation.\n",
      "tion.”\n",
      "lew\n",
      "beginning.\n",
      "worth.\n",
      "Wood.\n",
      "business.\n",
      "IbOo.\n",
      "l.\n",
      "Iowa..\n",
      "destruction\n",
      "to\n",
      "Interment\n",
      "a\n",
      "Country,\n",
      "says.\n",
      "these\n",
      "field.\n",
      "on.\n",
      ",\n",
      "water.\n",
      "gether.\n",
      "Trustees.\n",
      "office.\n",
      "said:\n",
      "arrived.-\n",
      "wares?\"\n",
      "operations.\n",
      "going.\n",
      "a\n",
      "thl\n",
      "consideration.\n",
      "r.\n",
      "are!\n",
      "done.\n",
      "rules.\n",
      "specimen,\"\n",
      "altetnoon.\n",
      "\"\n",
      "aud\n",
      "mentioned.\n",
      "How?\n",
      "cured.\"\n",
      "mention.\n",
      "Tur\n",
      "shrieking,\n",
      "And\n",
      "city.\n",
      "to-day.\n",
      "politics,\n",
      "an\n",
      "Newark.\n",
      "ordeal.\n",
      "primary.\n",
      "men,\"\n",
      "d\n",
      "them.\n",
      "...\n",
      "wit:\n",
      "?\n",
      "mind\n",
      "plasters.\n",
      "there.\n",
      "resource*.\n",
      "Yours,\n",
      "him.\n",
      "district\n",
      "ican.\n",
      "west.'''\n",
      "opportunities\n",
      "infinitum.\n",
      "by.”\n",
      "devices.\n",
      "health.\n",
      "course,\n",
      "voters.\n",
      "nlfi'clnd.\n",
      "a\n",
      "of\n",
      "ss\n",
      "Passaic\n",
      "Express.\n",
      "he\n",
      "wh\n",
      "prisonment.\n",
      "Trustees.\n",
      "j,\n",
      "appurtenances.\n",
      "p-\n",
      "color.\n",
      "tentiary.\n",
      "commerce.\n",
      "named.\n",
      "occurred.\n",
      "behalf.\n",
      "Science.\n",
      ".\n",
      "branch\n",
      "as\n",
      "attended.\n",
      "legislation.\n",
      "\"effective.\"\n",
      "sil\n",
      "I\n",
      "nt,\n",
      "luth.\n",
      "friends\n",
      "Hannah.\n",
      "Department.\n",
      "banks.\n",
      ",0c\n",
      "him*\n",
      "Edith,\n",
      "law.\n",
      "the\n",
      "their\n",
      "y,\n",
      "harm.\n",
      "ins.'\n",
      "nominated.\n",
      "answer.\n",
      "degree.\n",
      "drifted.\n",
      "Ostf-ta-\n",
      "nationalities.\n",
      "n.e\n",
      "n»A>..J!.\n",
      "sides.\n",
      "$2,250,000.\n",
      "lodge.\n",
      "described:\n",
      "Fos\n",
      "M.\n",
      "state\n",
      "e\n",
      "the\n",
      "or\n",
      "--\n",
      "da\n",
      "indigestion.\n",
      "continued:\n",
      "K\"venuuonl.\n",
      "court.\n",
      "tion.\n",
      "Kllevlts\n",
      "man.\n",
      "told.\n",
      "numb'--\n",
      "were\n",
      "carpet-baggers.-\n",
      "quality.\n",
      "on\n",
      "events\n",
      "more.\n",
      "'twos.\"\n",
      "Idler.\n",
      "roots.\n",
      "ginning.\n",
      "world\n",
      "in\n",
      "remedv.\n",
      ">.\n",
      "Bi.ll.\n",
      "Presldeot-\n",
      "died?\"\n",
      "supervisors.\n",
      "Cascades.\n",
      "twelve\n",
      "Atlantic.\n",
      "Glentworth.\n",
      "nationality.\n",
      "up?\n",
      "r(\n",
      "one.\n",
      "to-morrow.\n",
      "News.\n",
      "vance.\n",
      "e\n",
      "tustc.\n",
      "Herald.\n",
      "show.\n",
      "the\n",
      "de-\n",
      "Bottomley.\n",
      "estate.\n",
      "bargain.\n",
      "He\n",
      "list.\n",
      "strikes,\n",
      "No.\"\n",
      "beginning.\n",
      "mio-\n",
      "did.\n",
      "the\n",
      "1860.\n",
      "right.\n",
      "renown.\n",
      "co.,\n",
      "relense.\n",
      "Micbigau.\n",
      "5\n",
      "country.\n",
      "tor\n",
      "house.\"\n",
      "avenue.\n",
      "conference.\n",
      "o\n",
      "undischarged.\n",
      "gamuts.\n",
      "time.\n",
      "supporters.\n",
      "copy.\n",
      "conipauies.\n",
      "minutet.\"\n",
      "Number.\n",
      "the\n",
      "again,\n",
      "lea.\n",
      "1:474.\n",
      "regretted\n",
      "thence\n",
      "district.\n",
      "cation.\n",
      "d\n",
      "statements.\n",
      "Lyra,\n",
      "short-live-\n",
      "f\n",
      "s-.\n",
      "a-\n",
      "cats.\n",
      "d,\n",
      "product,\n",
      "work.\n",
      "absurdity\n",
      "SOft.\n",
      "city.\n",
      "Times.\n",
      "the\n",
      "hut.\n",
      "Veterans.\n",
      "income!\n",
      "y\n",
      "year.\n",
      "minister.\n",
      "filing.\n",
      "Press.\n",
      "them.\n",
      "sad\n",
      "S.\n",
      "to!d\n",
      "known.\n",
      "to.\"\n",
      "often\n",
      "tbe\n",
      "dealers.\n",
      "Jcnn\n",
      "less.\n",
      "pression.\n",
      "expected.\n",
      "members\n",
      "sustain\n",
      "r-\n",
      "claimants.\n",
      "P.\n",
      "order.\n",
      "had\n",
      "'\n",
      "seventies.\n",
      "authorizing\n",
      "notice.\n",
      "self-government.\n",
      "Street.\n",
      "redressed.\n",
      ",\n",
      "Appleby\n",
      "seed.\n",
      "places.\n",
      "wisdom.\n",
      "removed.\n",
      "kinds\n",
      "Virginia.\n",
      "7a9%c.\n",
      "away.\n",
      "tin\n",
      "Center.\n",
      "clerk.\n",
      "loving,\n",
      "y\n",
      "now.\n",
      "it\n",
      "continued.\n",
      "tli\n",
      "vision,\n",
      "Int.\n",
      "Investigations,\n",
      "the\n",
      "nillllu.\"\n",
      "b-\n",
      "pioneer.\n",
      "ed.\n",
      "Tobeauroliolathoh\n",
      "pjndlcutc.\n",
      "others.\n",
      "cusable.\n",
      "other\n",
      "con*\n",
      "stomachs.\"\n",
      "borer.\n",
      "ships.\n",
      "are\n",
      "youngsters.\n",
      "the-\n",
      "Record.\n",
      "Tiki\n",
      "on.—Southron\n",
      "complaint.\n",
      "\\\\\n",
      "her.\n",
      "remedy.\n",
      "neighbors.\n",
      "own\n",
      "children.\"\n",
      "Tobolsk.\n",
      "our\n",
      "U.I\n",
      "from.\n",
      "centuries.\n",
      "pigeons.\n",
      "Brown's.\n",
      "emetic.\n",
      "plays.\n",
      "disclosed.\n",
      "itr\n",
      "money.\n",
      "ways.\n",
      "discovered,\n",
      "man\n",
      "Iniallmetita\n",
      "calls.\n",
      "election.\n",
      "exclaiming.\n",
      "BarrlngtonBrown\n",
      "said,\n",
      "was\n",
      "nityasafeme’\n",
      "\"Gee,\n",
      "lazy.\n",
      "story.\n",
      "era.\"\n",
      "again,\n",
      "Harrl-\n",
      "Sim.\n",
      "walls.\n",
      "Jr.,\n",
      "scandal.\n",
      "ers\n",
      "ringing.\n",
      "police.\n",
      "ones.\n",
      "the\n",
      "er\n",
      "nence.\n",
      "them.\n",
      "tr\n",
      "trap.\n",
      "people.\n",
      ".\n",
      "hlr\n",
      "ployment.\n",
      "hills.\n",
      "•\n",
      "aforesaid\n",
      "E.VzS.E.\n",
      "1917.\n",
      "it.\n",
      "Thurmont\n",
      "harm.\n",
      "afford.\n",
      "life.\n",
      "moments.\n",
      "home!”\n",
      "purpose.\"\n",
      "earth.\n",
      "purpose.\n",
      "In.\n",
      "While\n",
      "perlenced.\n",
      "salo.\n",
      "of.\n",
      "it.\n",
      "on\n",
      "minimum.\"\n",
      "Burgess)\n",
      "Territory.\n",
      "Lothian\n",
      "hand.\n",
      "life.\n",
      "yachtsmen\n",
      "taken.\n",
      "person.\n",
      "M.\n",
      "motions.\n",
      "rivalry\n",
      ",\n",
      "other.\n",
      "ous\n",
      "prosjierity.\n",
      "Abbott.\n",
      "gone.\n",
      "claims.\n",
      "morality.\n",
      "de\n",
      "Co.\n",
      "foothills.\n",
      "ult.\n",
      "Proctor.\n",
      "ii.\n",
      "End.\n",
      "heed.\n",
      "T\n",
      "Commissioner\n",
      ".\n",
      "others.\n",
      "able-bodi-\n",
      "lOWL.\n",
      "filing.\n",
      "transfer.\n",
      "pig.\n",
      "life\n",
      "1,000.\n",
      "g\n",
      "crime.\n",
      "brothers.\n",
      "atTVVc\n",
      "the\n",
      "little.\n",
      "dc-\n",
      "s\n",
      "he\n",
      "cure.\n",
      "fund.\n",
      "of\n",
      "manner\n",
      "7\n",
      "Post.\n",
      "avoided.\n",
      "it.\n",
      "over.\n",
      "Advocate\n",
      "population.”\n",
      "box.\n",
      "Patriot.\n",
      "in\n",
      "failure.\n",
      "backers.\n",
      "affection.\n",
      "Office.\n",
      "have.\n",
      "--\n",
      "setting.\n",
      "ailment.\n",
      "long.\n",
      "ta\n",
      "justified.\n",
      "Con\n",
      "In\n",
      "taW\n",
      "yet.\n",
      "fruit.\n",
      "diseases.\n",
      "exposed.\n",
      "them.\n",
      "servant,\n",
      "geiu'ratlou.\n",
      "Justice.\n",
      "spotless,\n",
      "air.\n",
      "this.\"\n",
      "silver\n",
      "L................................\n",
      "needed.\n",
      "rate.\n",
      "Intelligencer.\n",
      "have\n",
      "aostroyed.\n",
      "the\n",
      "ten\n",
      "meanness.”\n",
      "at\n",
      "figures.\n",
      "strawberries,\n",
      "follow\n",
      "system\n",
      "remain\n",
      "man.\n",
      "prosecutions\n",
      "registry-\n",
      ".\"\n",
      "Hevlew.\n",
      "is\n",
      "of\n",
      "ge-\n",
      "alone.\n",
      "uaryL\n",
      "the\n",
      "him.\n",
      "n\n",
      "like\n",
      "artists.\n",
      "Americans.”\n",
      "interest:\n",
      "Delaware.\n",
      "loa's.\n",
      "fering.\n",
      "winner.\"\n",
      "look.\n",
      "up.\n",
      "California.\n",
      ".\n",
      "the\n",
      "respect.\n",
      "fed.\n",
      "Lyman.\n",
      "quiredi\n",
      "-\n",
      "tobacco\n",
      "height\n",
      "dinner.\"\n",
      "tors.\n",
      "basin.\n",
      "$400.-\n",
      "ad-\n",
      "Chronicle.\n",
      "practices.\n",
      "rality.\n",
      "-\n",
      "people.\n",
      "believe\n",
      ".\n",
      "decline.\n",
      "Liverpool,\n",
      "onlj\n",
      "sort.\n",
      "course.\n",
      "Vou\n",
      "In\n",
      "why.\n",
      "since.\n",
      "plication.\n",
      "Tesreau.\n",
      "association.\n",
      "-\n",
      "parlors.\n",
      "across\n",
      "tippurunning.\n",
      "the\n",
      "Post.\n",
      "1880.\n",
      "lows:\n",
      "party.\n",
      "bejth.\n",
      "now.\n",
      "10CO\n",
      "fulminations.\n",
      "says:\n",
      "ago.\n",
      "employ.\n",
      ".\n",
      "haste.\n",
      "dispensation.\n",
      "might\n",
      "the\n",
      "seat\n",
      "erful\n",
      "convenience\n",
      "tion.\n",
      "possessed.\n",
      "write.\n",
      "lata.\n",
      "was-no-\n",
      "miner.\n",
      "g\n",
      "replied.\n",
      "handsome.\n",
      "th\"ir\n",
      "enter.\n",
      "i\n",
      "that?\n",
      "r.\n",
      "both.\n",
      "inherited\n",
      "place.\n",
      "g\n",
      "ccptance.\n",
      "hue.\n",
      "nope\n",
      "my\n",
      "heaven.\n",
      "feet.\n",
      "Egj-pt-\n",
      "from\n",
      "another.\n",
      "baggers.\n",
      "75(380c.\n",
      "to\n",
      "Causey,\n",
      "^\n",
      "13.\n",
      "most\n",
      "destiny-\n",
      "week\n",
      "the\n",
      "?\n",
      "principal.\n",
      "drugs.\n",
      "14-th.\n",
      "do.\n",
      "tii\n",
      "service.\n",
      "tives.\n",
      "follows\n",
      "Darien.\n",
      "Ui\n",
      "y\n",
      "Benwood.\n",
      "greenbacks.\n",
      "artillery.\n",
      "died.\n",
      "R«ncy.\n",
      "tube.\n",
      "once.'\n",
      "high.\"\n",
      "house.\"\n",
      "pcuuiles\n",
      "explorer.\n",
      "Appeal.\n",
      "development.\n",
      "game?’’\n",
      "Journal.\n",
      "him.”\n",
      "^y\n",
      "the\n",
      "Amerlrnn.\n",
      "lands.\n",
      "jl\n",
      "HI\"\n",
      "came\n",
      "customers.\n",
      "aforesaid.\n",
      "them.\n",
      "them.\n",
      "Matthews.\n",
      "Babbitt.\"\n",
      "officers.\n",
      "show.\n",
      "pursuit.\n",
      "there\n",
      "it.\n",
      "ever.\n",
      "opin-\n",
      "fnrnishlnj-\n",
      "haud.\n",
      "Solnt.\n",
      "iu\n",
      "payments.\n",
      "wage*.\n",
      "lawlessness.\n",
      "removed.\n",
      "census-taker-s\n",
      "worst--\n",
      "today',\n",
      "captivity.\n",
      "Ot.\n",
      "sum.\n",
      "is.\"\n",
      "therefore,\n",
      "up\n",
      "dimensions.\n",
      "spot.\n",
      "of\n",
      "Celluloid.\n",
      "tion,\n",
      "fools.\n",
      "Saturdays.\n",
      "attention.\n",
      "something\n",
      "forks.\n",
      "amy.\n",
      "?\n",
      "har­\n",
      "Medical\n",
      "living\n",
      "perienced.\n",
      "er\n",
      "hut\n",
      "istrar*.\n",
      "of\n",
      "expo-\n",
      "t\n",
      "New-Cen-\n",
      "nnuoyc-d-\n",
      "dry\n",
      "scandal.\"\n",
      "pass.\n",
      "A\n",
      "make\n",
      "President.\n",
      "team.\n",
      "ts\n",
      "s\n",
      "discussion.\n",
      "MiningCompan-,\n",
      "M»5.\n",
      "‘ltion\n",
      "be-\n",
      "Asia\n",
      "Thejr\n",
      "him.\n",
      "and\n",
      "fine-looki-\n",
      "Involved.\n",
      "land.\n",
      "weeks.\n",
      "State.\n",
      "missionary\n",
      "heart.\n",
      "yours,\n",
      "liberty.\n",
      "g\"\n",
      "life-an-\n",
      "anew.\n",
      "line.!\n",
      "ninjf.\n",
      "day.\n",
      "that\n",
      "my\n",
      "logs.'\n",
      "Tress.\n",
      "matter.\n",
      "they\n",
      "a\n",
      "wires\n",
      "months.\n",
      "¬\n",
      "SECRETS,\n",
      "recover.\n",
      "ifian.\n",
      "admin-\n",
      "isa\n",
      "consequence.\n",
      "50.\n",
      ",\n",
      "for\n",
      "pray.\n",
      "banner.\n",
      "those\n",
      "Beecher,\n",
      "“\n",
      "oi\n",
      "ph.\"\n",
      "says:\n",
      ".\n",
      "commissioners.\n",
      "repair.\n",
      "conjecture.\n",
      "-\n",
      "ley.\n",
      "Duffy,\n",
      "time.\n",
      "evenings.\n",
      "shot\n",
      "boor.\n",
      "operate\n",
      "\"gibbet,\"\n",
      "ing.\n",
      "murdoror.\n",
      "theud-\n",
      "°l\n",
      "nomination.\n",
      "unflinchingly.\n",
      "suddenness.\n",
      "aimnch\n",
      "cup.\n",
      "bnJge.\n",
      "road.\n",
      "clients.\n",
      "division.\n",
      "hat\n",
      "-\n",
      "tirm.\n",
      "Age,\n",
      "tunate,\n",
      "ju-Jifice.\n",
      "be\n",
      "particular.\n",
      "enterprise.\n",
      "tl..-\n",
      "Bhortatl.\n",
      "influence.\n",
      "preserves.\n",
      "-\n",
      "vlllago\n",
      "respon-dbl-\n",
      "survivors.\n",
      "8,000.\n",
      "position.\n",
      "Railroad-:\n",
      "Khln.\n",
      "printed.\n",
      "State,\n",
      "eat.\n",
      "appertaining.\n",
      "match.\n",
      "Ii\n",
      "mines.\n",
      "or\n",
      "them.”\n",
      "afternoons.\n",
      "directed\n",
      "claims.\n",
      "d,\n",
      "D.\n",
      "with\n",
      "demanded.\n",
      "Brown,\n",
      "for\n",
      "iron.\n",
      "law.\n",
      "out\n",
      "interest.\n",
      "44\n",
      ".1\n",
      "Donehoo.\n",
      "onto-\n",
      "health.\n",
      "labor.'\n",
      "o:\n",
      "vehicles.\n",
      "II...\n",
      "solid.\n",
      "term.\n",
      "Germany.\n",
      "telling\n",
      "church.\n",
      "inoffeusive.\n",
      "performed,\n",
      "folly.\n",
      "outing.\n",
      "gency.\n",
      "within.\n",
      "of\n",
      "Hanker.\n",
      "leaders.\n",
      "be,\n",
      "car.\n",
      "then\n",
      "made.\n",
      "enco.\n",
      "sunshine.\n",
      "Hamilton.\n",
      "Prospect\n",
      "lay.\n",
      "aueatinn\n",
      "cneap.\n",
      "member­\n",
      "Catholio-ity-\n",
      "peace.\n",
      "air.\n",
      "its\n",
      "preachers.\"\n",
      "license.\n",
      "wick,\"\n",
      "blight.\n",
      "of\n",
      "Duncan,\n",
      "provisions.\"\n",
      "in\n",
      "on.\n",
      "reclamation\n",
      "Jnly,\n",
      "made.\n",
      "joke?\n",
      "Agency.\n",
      "clut-\n",
      "-poli-\n",
      "passage.\n",
      "system.\n",
      "|kj1.uo\n",
      "troops.”\n",
      "work.\n",
      "estab­\n",
      "disease.\n",
      "y\n",
      "vots.\n",
      "yet.\n",
      "regular\n",
      "Ptttl\n",
      "murder.\n",
      "lap.\n",
      "Science.\n",
      "gentleman.\n",
      "tions.\n",
      "anywhere.\n",
      "therf-\n",
      "Congress.\n",
      "reduction\n",
      "of\n",
      "Senators.\n",
      "C;\n",
      "garchy.\n",
      "lunatics.\n",
      "sale.\n",
      "work.\n",
      "1860.\n",
      "Toledo.\n",
      "-I\n",
      "her.\n",
      "Magazine.\n",
      "mind.\n",
      "State.\n",
      "McClurc.\n",
      "view.\n",
      "It.\n",
      "laid\n",
      "to-day.\n",
      "coals.\"\n",
      "health.”\n",
      "forfeited.\n",
      "people.\n",
      "sioner.\n",
      "tion.\n",
      "Tehtiacan.\n",
      "jear.\n",
      "provisions.\n",
      "prostration.\n",
      "mented\n",
      "neau.\n",
      "rates.\n",
      "TTVe-\n",
      "properties.”\n",
      "is\n",
      "favo.-\n",
      "Philadelphia.\n",
      "ing\n",
      "prices.\n",
      "accident.\n",
      "send\n",
      "music.\n",
      "$1200.\n",
      "him.\n",
      "appetites\n",
      "shoes.\n",
      "existence.\n",
      "beginning.\n",
      "fact.\n",
      "food.\n",
      "transportation.\n",
      "forever!\n",
      "otTer.\n",
      "Plalnville-Farmingto-\n",
      "3--\n",
      "collision.\n",
      "sentiment,\n",
      "At^et\n",
      "anoth\n",
      "direction.\n",
      "reduction,\n",
      "service.\n",
      "America,-\n",
      "health.\n",
      "sums.\n",
      "good.\"\n",
      "Plant\n",
      "thing.\n",
      "News.\n",
      "no\n",
      "Washington.\n",
      "Times.\n",
      "Democratic\n",
      "prevailed.\n",
      "agreement.\n",
      "specially\n",
      "allowed.\n",
      "we\n",
      "treated.\n",
      "d«ir\n",
      "connection.\n",
      "irvnn.\n",
      "offer.\n",
      "ways.\n",
      "st\n",
      "summit.\n",
      "men.\n",
      "trade.\n",
      "injured.\n",
      "branch\n",
      "case.\n",
      "night.\n",
      "ibute.\n",
      "e\n",
      "lf\n",
      "power.\n",
      "simple\n",
      "ths\n",
      "world.''\n",
      "ini-\n",
      "with\n",
      ".\n",
      "surface.\"\n",
      "men.\"\n",
      "Nome.\n",
      "things.\"\n",
      "Hie\n",
      "Mountain.\n",
      "1858.\n",
      "wishes\n",
      "damages.\n",
      "settlement.\"\n",
      "________\n",
      "Y.\n",
      "sermons.\n",
      "costs.\n",
      "contain.\n",
      "uncertain.\n",
      "delay.\n",
      "cheese\n",
      "we\n",
      "A\n",
      "homes.\n",
      "-\n",
      "button.\n",
      "fate.\n",
      "n\n",
      "district.\n",
      "up.\n",
      "said:\n",
      "other.\n",
      "latter.\n",
      "lat*\n",
      "de-\n",
      "c\n",
      "off.\"\n",
      "1874.\"\n",
      "t.\n",
      "on\n",
      "the-\n",
      "proposed.\n",
      "#\n",
      "distance.\n",
      "will.\n",
      "rooms.\n",
      "paid\n",
      "suggest-th-\n",
      "oligarchy.\n",
      "sea.\n",
      "rt\n",
      "eleven\n",
      "tion.\n",
      "recent\n",
      "information.\n",
      "otllcliils.\n",
      "shock.\n",
      "home.\n",
      "Kensington.\n",
      "heart.\n",
      "by\n",
      "place.\n",
      "Equalization.\n",
      "Parker.\n",
      "come.\n",
      "upon.\n",
      "yours,\n",
      "placed\n",
      "and\n",
      "elections.\n",
      "tices.\n",
      "Louis,\n",
      "supply.\n",
      "Taft.\n",
      "days.\n",
      "Tuttle\n",
      "common\n",
      "voices\n",
      ".\n",
      "avail.\n",
      "siasm.\n",
      "<\n",
      "sup*\n",
      "navies\n",
      "things.\"\n",
      "safety.\n",
      "my-\n",
      "-\n",
      "follows:\n",
      "rub.\n",
      "effected.\n",
      "nation.”\n",
      "her.\n",
      "street,\n",
      "ena-\n",
      "date.\n",
      "nnd\n",
      "Informants.\n",
      "Porter,\n",
      "mad.\n",
      "vengeance.\n",
      "l._\n",
      "you.\n",
      "railroad\n",
      "iuy\n",
      "considered.\n",
      "storm.\n",
      "qualities.\n",
      "Monday.\n",
      "let-\n",
      "and\n",
      "Newark.\n",
      "at-\n",
      "vote.\n",
      "all.\n",
      "industries.\n",
      "ama.\n",
      "nations,\n",
      "e\n",
      "city.\n",
      "the.GMeeï\n",
      "don't\n",
      "w\n",
      "started\n",
      "effected.\n",
      "Ohio.\n",
      "systun.\n",
      "i\n",
      "armies\n",
      "and\n",
      "family.\n",
      "Is\n",
      "sleep.\n",
      "Bazar.\n",
      "the\n",
      "Lodge,\n",
      "consumption.\n",
      "no\n",
      "St\n",
      "puzzler.\n",
      "repaired.\n",
      "\"\n",
      "organs\n",
      "long\n",
      "time.\n",
      "important.\n",
      "terms.\n",
      "n-\n",
      "can\n",
      "\\\\\n",
      "his\n",
      "som--\n",
      "street.\n",
      "a\n",
      "law.\n",
      ";\n",
      "pa;\n",
      "\"setback.\"\n",
      "waters.\n",
      "naked\n",
      "unskilled\n",
      "generally.\n",
      "bier.\"\n",
      "Us.”\n",
      "land\n",
      "rroiirl«torofferi«.\n",
      "values.\n",
      "left.\n",
      "3\n",
      "file.\n",
      "observed.''\n",
      "ring.\n",
      "president.\n",
      "i3.\n",
      "them.\n",
      "wbicn\n",
      "year.\n",
      "us.\n",
      "phalanx\n",
      "bow.\n",
      "j\n",
      "afterward.\"\n",
      "Liens.\n",
      "Congress.\n",
      "yore.\n",
      ":\n",
      "HUH.\n",
      "ger.\n",
      "intellectual\n",
      "onions.\n",
      "possible.'\n",
      "be.\n",
      "»SS».\n",
      "\"\n",
      "Titus.\n",
      "conspirators.\n",
      "toys,\n",
      "de-\n",
      "7414\n"
     ]
    }
   ],
   "source": [
    "\n",
    "with lzma.open(test_file, 'rt') as file:\n",
    "    predict_words = []\n",
    "    results = []\n",
    "    for line in file:\n",
    "#         print(line)\n",
    "        line = preprocess(line) #get only relevant\n",
    "        split = line.split('\\t')\n",
    "        print(get_last_word(split[0]))\n",
    "        predict_words.append(get_last_word(split[0])) #get_first_word(split[1])\n",
    "    print(len(predict_words))\n",
    "    vocab = train_dataset.vocab\n",
    "    for presc_word in predict_words:\n",
    "        results.append(dict(get_values_from_model(presc_word, model, vocab, k=k)))\n",
    "    \n",
    "    \n",
    "    with open(out_file, 'w') as outfile:\n",
    "        for elem in results:\n",
    "            \n",
    "            outfile.write(gonito_format(elem))\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.6"
  },
  "org": null
 },
 "nbformat": 4,
 "nbformat_minor": 1
}