{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "import torch \n",
    "import pandas\n",
    "import numpy\n",
    "from torch.autograd import Variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "class LinearRegressionModel(torch.nn.Module): \n",
    "    def __init__(self): \n",
    "        super(LinearRegressionModel, self).__init__() \n",
    "        self.linear = torch.nn.Linear(1, 1)\n",
    "    def forward(self, x): \n",
    "        y_pred = self.linear(x) \n",
    "        return y_pred "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_data=pandas.read_csv('C:/Users/eryk6/PycharmProjects/mieszkania5/train/train.tsv',sep='\\t',header=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "x=train_data[0].tolist()\n",
    "x = [str(train_data).replace(' ', '') for train_data in x]\n",
    "\n",
    "y=train_data[8].tolist()\n",
    "y = [str(train_data).replace(' ', '') for train_data in y]\n",
    "\n",
    "x=numpy.array(x, dtype=numpy.float32)\n",
    "y=numpy.array(y, dtype=numpy.float32)\n",
    "\n",
    "x = x.reshape(-1, 1)\n",
    "y = y.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = LinearRegressionModel()\n",
    "criterion = torch.nn.MSELoss() \n",
    "optimizer = torch.optim.SGD(our_model.parameters(), lr = 0.000000000001 ) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "for i in range(500): \n",
    "    input = Variable(torch.from_numpy(x))\n",
    "    pred_y = model(input)\n",
    "    optimizer.zero_grad()\n",
    "    loss = criterion(pred_y, Variable(torch.from_numpy(y))) \n",
    "    optimizer.zero_grad() \n",
    "    loss.backward() \n",
    "    optimizer.step()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_A_in = pandas.read_csv('C:/Users/eryk6/PycharmProjects/mieszkania5/test-A/in.tsv',sep='\\t',header=None)\n",
    "x = test_A_in[7].tolist()\n",
    "x = numpy.array(x, dtype=numpy.float32)\n",
    "x = x.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = model(Variable(torch.from_numpy(x))).data.numpy()\n",
    "out = open('C:/Users/eryk6/PycharmProjects/mieszkania5/test-A/out.tsv', 'w')\n",
    "for i in y:\n",
    "    out.write(str(i[0])+'\\n')\n",
    "out.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "dv_in = pandas.read_csv('C:/Users/eryk6/PycharmProjects/mieszkania5/dev-0/in.tsv',sep='\\t',header=None)\n",
    "x = dv_in[7].tolist()\n",
    "x = numpy.array(x, dtype=numpy.float32)\n",
    "x = x.reshape(-1, 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = model(Variable(torch.from_numpy(x))).data.numpy()\n",
    "output = open('C:/Users/eryk6/PycharmProjects/mieszkania5/dev-0/out.tsv', 'w')\n",
    "for i in y:\n",
    "    output.write(str(i[0])+'\\n')\n",
    "output.close()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3.6 (tensorflow)",
   "language": "python",
   "name": "tensorflow"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}