{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imports"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pickle\n",
    "from gensim.models import Word2Vec"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Configuration\n",
    "\n",
    "*models_filename:* The complete path to the pickeld word2vec models.\n",
    "\n",
    "*word_pairs_filename:* The complete path to the list of word pairs used for evaluation. This needs to be a **.csv** file.\n",
    "\n",
    "*selected_model_filename*: The filename for the best performing model which will be used for the subsequent classification. You may use the **.p** extension indicating a pickled file, but you are free to use whatever you like."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "models_filename = \"models.p\"\n",
    "word_pairs_filename = \"ready_to_use/word_pairs/French.csv\"\n",
    "selected_model_filename = \"best_model.p\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Model Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "print(\"loading models\")\n",
    "with open(models_filename, \"rb\") as handle:\n",
    "    models = pickle.load(handle)\n",
    "print(\"loaded {} models\".format(len(models)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Evaluate models"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "best_sim = 0\n",
    "best_pars = None\n",
    "for p, m in models.items():\n",
    "    similarities = m.wv.evaluate_word_pairs(word_pairs_filename, delimiter=\",\")\n",
    "    if similarities[0][0] > best_sim:\n",
    "        best_sim = similarities[0][0]\n",
    "        best_pars = p\n",
    "print(\"found best model with parameters: {}\".format(best_pars))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Save best performing model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(selected_model_filename, \"wb\") as handle:\n",
    "    pickle.dump(models[best_pars], 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
}