kleister-nda/.ipynb_checkpoints/run-checkpoint.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "8b07c9a5-e5cf-4cf9-a6d9-e784eb109fef",
   "metadata": {},
   "outputs": [],
   "source": [
    "import re"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "fce94c21-6792-4938-bf2c-3f46ecf2f954",
   "metadata": {},
   "outputs": [],
   "source": [
    "states = ['Alabama', 'Alaska', 'Arizona', 'Arkansas', 'California', 'Colorado', 'Connecticut', 'Delaware', 'Florida', 'Georgia', \n",
    "        'Hawaii', 'Idaho', 'Illinois', 'Indiana', 'Kansas', 'Kentucky', 'Louisiana', 'Maine', 'Maryland', \n",
    "        'Massachusetts', 'Michigan', 'Minnesota', 'Mississippi', 'Missouri', 'Nebraska', 'Nevada', 'New Hampshire', 'New Jersey', \n",
    "        'New Mexico', 'New York', 'North Carolina', 'North Dakota', 'Ohio', 'Oklahoma', 'Pennsylvania', 'Rhode Island', 'South Carolina', \n",
    "        'South Dakota', 'Tennessee', 'Texas', 'Utah', 'Vermont', 'Virginia', 'Washington', 'West Virginia', 'Wisconsin', 'Wyoming']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "eb1815f2-1876-4437-833a-ff22de81685e",
   "metadata": {},
   "outputs": [],
   "source": [
    "def counter(text_in, query):\n",
    "    pattern = re.compile(query)\n",
    "    return len(pattern.findall(text_in, re.IGNORECASE))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "8729062d-87b8-4111-a216-8500334f54b6",
   "metadata": {},
   "outputs": [],
   "source": [
    "def state_prediction(text_in):\n",
    "    state_dict = {}\n",
    "    for state in states:\n",
    "        state_dict[state.replace(\" \", \"_\")] = counter(text_in, state)     \n",
    "    return max(state_dict, key=state_dict.get)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "ea8069f7-de8e-454c-8eac-9fb7cc0df626",
   "metadata": {},
   "outputs": [],
   "source": [
    "def jurisdiction(path_in, path_out):    \n",
    "    with open(path_in, 'r', encoding='utf8') as file:\n",
    "        lines = file.readlines()\n",
    "    with open(path_out, 'wt')as file_out:\n",
    "        for i in lines:\n",
    "            file_out.write(\"jurisdiction=\"+str(state_prediction(i))+'\\n')            \n",
    "    file_out.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "ade45bfb-9eaa-4b2b-bba1-6cfcaf0a9ce6",
   "metadata": {},
   "outputs": [],
   "source": [
    "jurisdiction('dev-0/in.tsv', 'dev-0/out.tsv')\n",
    "jurisdiction('train/in.tsv', 'train/out.tsv')\n",
    "jurisdiction('test-A/in.tsv', 'test-A/out.tsv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "594a25a9-a0ce-4de9-82c8-df50a4ecac39",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[NbConvertApp] Converting notebook run.ipynb to script\n",
      "[NbConvertApp] Writing 1605 bytes to run.py\n"
     ]
    }
   ],
   "source": [
    "!jupyter nbconvert --to script run.ipynb"
   ]
  }
 ],
 "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": "3.9.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}