{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import glob\n",
    "import pickle\n",
    "import re\n",
    "from gensim.models import Word2Vec\n",
    "from gensim.models.phrases import Phrases, Phraser\n",
    "from itertools import product\n",
    "from stop_words import get_stop_words\n",
    "from nltk import tokenize"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Configuration\n",
    "\n",
    "*input_dir:* The path to the directory that contains your text files. Please make sure to use a '/' (slash) in the end. For example: path/to/texts/.\n",
    "\n",
    "*language*: The language of your texts. This is used to get the right list of stops words.\n",
    "\n",
    "*num_processes*: The number of processes to use. This depends on your hardware. The more cores you can use, the faster the training of the models.\n",
    "\n",
    "*models_filename:* The filename for the resulting trained models. You may use the **.p** extension indicating a pickled file, but you are free to use whatever you like. Just make sure this is consistent in the evaluation step."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "input_dir = \"../data/texts/\"\n",
    "language = \"french\"\n",
    "num_processes = 2\n",
    "models_filename = \"models.p\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Grid search parameters\n",
    "You should provide possible values for hyperparameters of the word2vec model. Please refer to the [gensim documentation](https://radimrehurek.com/gensim/models/word2vec.html#gensim.models.word2vec.Word2Vec) to see a list of all hyperparameter. The following values serve as an example and may be adjusted to your needs."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "vector_sizes = [100, 200, 300]\n",
    "skip_grams = [0, 1]\n",
    "hs = [0, 1]\n",
    "windows = [5, 10]\n",
    "negatives = [5, 10]\n",
    "iters = [5, 10]\n",
    "\n",
    "hyperparameters = list(product(vector_sizes, skip_grams, hs, windows, negatives, iters))\n",
    "num_hyperparameters = len(hyperparameters)\n",
    "print(\"number of hyperparameter combinations:\", num_hyperparameters)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Gird Search"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Loading texts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "text_file_names = glob.glob(\"{}*.txt\".format(input_dir))\n",
    "print(\"found {} texts\".format(len(text_file_names)))\n",
    "texts = []\n",
    "for text_file_name in text_file_names:\n",
    "    with open(text_file_name, \"r\", encoding=\"utf-8\") as input_file:\n",
    "        texts.append(input_file.read())\n",
    "print(\"loaded {} texts\".format(len(texts)))\n",
    "combined_text = \" \".join(texts) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conduct grid search"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "models = {}\n",
    "stop_words = get_stop_words(language.lower())\n",
    "sentences = tokenize.sent_tokenize(combined_text)\n",
    "reg_exp_tok = tokenize.RegexpTokenizer(r\"\\w{3,}\")\n",
    "split_sentences = [reg_exp_tok.tokenize(s.lower()) for s in sentences]\n",
    "split_sentences_wo_sw = []\n",
    "for s in split_sentences:\n",
    "    cleaned_tokens = [t for t in s if t not in stop_words]\n",
    "    if len(cleaned_tokens) > 0:\n",
    "        split_sentences_wo_sw.append(cleaned_tokens)\n",
    "for hp in hyperparameters:\n",
    "    model = Word2Vec(sentences=split_sentences_wo_sw, workers=num_processes, size=hp[0], sg=hp[1], hs=hp[2], window=hp[3], negative=hp[4], iter=hp[5])\n",
    "    models[hp] = model"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Save models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(models_filename, \"wb\") as handle:\n",
    "    pickle.dump(models, handle)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "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": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}