notebooks/model_logistic_regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "d21dc15f-e474-4b63-8337-21cae7b84e51",
   "metadata": {},
   "source": [
    "# 1. Setup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [
    "# add root path to system path\n",
    "import sys\n",
    "import os\n",
    "\n",
    "sys.path.append('../')\n",
    "\n",
    "os.chdir(\"../\")"
   ],
   "metadata": {
    "collapsed": false
   },
   "id": "987d27546675696c"
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2a164ea1-c512-422e-b840-e383ae8e0852",
   "metadata": {},
   "outputs": [],
   "source": [
    "import time\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfVectorizer\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.pipeline import Pipeline\n",
    "\n",
    "from utility.utils import export_submission, WordEmbeddingVectorizer, load_preprocessed_data, \\\n",
    "    load_preprocessed_full_data, load_preprocessed_test_data\n",
    "from utility.paths import DataPath\n",
    "\n",
    "import gensim"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e528b5a3-cb3b-4f18-aef7-d0176d894ec6",
   "metadata": {},
   "source": [
    "This notebook is similar to the one about [Stochastic Gradient Descent (SGD)](model_SGD.ipynb). Therefore, we won't go over the concepts again, but will instead focus on the results and discuss them."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b7f559a6-228f-4063-a7b7-a504d350f7f8",
   "metadata": {
    "tags": []
   },
   "source": [
    "# 2. Data loading"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fef0f06d-439a-4477-9ddb-ca3bcaaeb713",
   "metadata": {},
   "source": [
    "First, we will load the pre-processed datasets."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "10d344d4-94fa-4e15-9297-9c73d20b1d94",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the data\n",
    "data_train = load_preprocessed_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "fba80a18-7fbc-44fc-b86c-f8b47c22f34a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split the data into training and testing sets\n",
    "X_train, X_validation, y_train, y_validation = train_test_split(data_train[\"text\"], data_train['label'], test_size=0.2,\n",
    "                                                                random_state=42)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "00f31caf-e6b0-4d02-bfc9-5812dc2c7452",
   "metadata": {
    "tags": []
   },
   "source": [
    "# 3. Text Vectorizer"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73a3b502-3f03-4803-8a30-4034f380947b",
   "metadata": {},
   "source": [
    "Now, we will conduct a grid search to identify the optimal text vectorizer for **Logistic Regression** the followings :\n",
    "\n",
    "- `Count Vectorizer`\n",
    "- `TF-IDF Vectorizer`\n",
    "- `Word2Vec`\n",
    "\n",
    "This will include determining their best hyperparameters."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f0880803-0a15-4b66-b00a-8e1d12235d57",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### Count Vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "14d287eb-3c66-4682-97e8-b341ce6980d1",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Count Vectorizer - Best parameters: {'vect__ngram_range': (1, 3)}\n",
      "Count Vectorizer - Best score: 0.8155027999920159\n"
     ]
    }
   ],
   "source": [
    "# Define pipeline for Count Vectorizer\n",
    "pipeline_count = Pipeline([\n",
    "    ('vect', CountVectorizer()),\n",
    "    ('clf', LogisticRegression(max_iter=1000))\n",
    "])\n",
    "\n",
    "# Parameter grid for Count Vectorizer pipeline\n",
    "parameter_grid_count = {\n",
    "    'vect__ngram_range': [(1, 1), (1, 2), (1, 3)],\n",
    "}\n",
    "\n",
    "# Grid search for Count Vectorizer pipeline\n",
    "grid_search_count = GridSearchCV(pipeline_count, parameter_grid_count, cv=5)\n",
    "\n",
    "# Fit the grid search to the data\n",
    "grid_search_count.fit(X_train, y_train)\n",
    "\n",
    "# After fitting, we can access the best parameters, best score, etc.\n",
    "print(\"Count Vectorizer - Best parameters:\", grid_search_count.best_params_)\n",
    "print(\"Count Vectorizer - Best score:\", grid_search_count.best_score_)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a7a14adb-df2f-4a73-942b-e41aa473e124",
   "metadata": {
    "tags": []
   },
   "source": [
    "#### TF-IDF Vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "816306ec-d4e8-4a22-bd83-77e8633bcd57",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TF-IDF - Best parameters: {'vect__ngram_range': (1, 2)}\n",
      "TF-IDF - Best score: 0.8068081700760263\n"
     ]
    }
   ],
   "source": [
    "# Define pipeline for TF-IDF\n",
    "pipeline_tfidf = Pipeline([\n",
    "    ('vect', TfidfVectorizer()),\n",
    "    ('clf', LogisticRegression(max_iter=1000))\n",
    "])\n",
    "\n",
    "# Parameter grid for TF-IDF pipeline\n",
    "parameter_grid_tfidf = {\n",
    "    'vect__ngram_range': [(1, 1), (1, 2), (1, 3)],\n",
    "}\n",
    "\n",
    "# Grid search for TF-IDF pipeline\n",
    "grid_search_tfidf = GridSearchCV(pipeline_tfidf, parameter_grid_tfidf, cv=5)\n",
    "\n",
    "# Fit the grid search to the data\n",
    "grid_search_tfidf.fit(X_train, y_train)\n",
    "\n",
    "# After fitting, you can access the best parameters, best score, etc.\n",
    "print(\"TF-IDF - Best parameters:\", grid_search_tfidf.best_params_)\n",
    "print(\"TF-IDF - Best score:\", grid_search_tfidf.best_score_)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3efae94c-b835-438d-9c2a-ad0b0a3cdfcc",
   "metadata": {},
   "source": [
    "#### Word2Vec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "6796bf75-c2d4-4606-b632-256a6ba69f55",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load Google's pre-trained Word2Vec model.\n",
    "model_path = 'data/GoogleNews-vectors-negative300.bin'\n",
    "word_vectors = gensim.models.KeyedVectors.load_word2vec_format(model_path, binary=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "66e04609-48c4-404f-951c-28d3002e51c1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Accuracy: 0.6272822549506316\n"
     ]
    }
   ],
   "source": [
    "# Create an instance of the WordEmbeddingVectorizer with our Word2Vec model\n",
    "word_vectorizer = WordEmbeddingVectorizer(word_vectors)\n",
    "\n",
    "# Create a pipeline\n",
    "pipeline = Pipeline([\n",
    "    ('word2vec', word_vectorizer),\n",
    "    ('classifier', LogisticRegression(max_iter=1000))\n",
    "])\n",
    "\n",
    "# Train the model\n",
    "pipeline.fit(X_train, y_train)\n",
    "\n",
    "# Calculate and print the accuracy\n",
    "accuracy = pipeline.score(X_validation, y_validation)\n",
    "print(\"Accuracy:\", accuracy)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff3bafdd-2f49-4663-bba7-454283eb96c4",
   "metadata": {
    "tags": []
   },
   "source": [
    "# 4. Optimization of the Logistic Regression model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ee552d20-1cd0-4ac5-9a17-ae6916672b86",
   "metadata": {},
   "source": [
    "As with the SGD model, the vectorizer that yields the best accuracy is `CountVectorizer` with the parameters `ngram_range=(1, 3)`. Now let's use it to find the best Logistic Regression hyperparameters with a grid search."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "330a6c8c-2244-4fa8-be0e-38d2354deebd",
   "metadata": {},
   "source": [
    "### Definition of the model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2d9c8b1-4b28-4c9a-a020-40a735cc980e",
   "metadata": {},
   "source": [
    "First, let's vectorize the data with the best vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "b2fb8b88-274b-409c-9591-9384d29faab4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# CountVectorizer\n",
    "best_vectorizer = CountVectorizer(ngram_range=(1, 3))\n",
    "X_train_vectorized = best_vectorizer.fit_transform(X_train)\n",
    "X_validation_vectorized = best_vectorizer.transform(X_validation)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6ae2fea3-63a9-4980-94eb-503413de4a08",
   "metadata": {},
   "source": [
    "Now let's train the Logisitic Regression model with the vectorized data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "859c8e68-df99-45b0-b412-84490e5233fc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training time: 141.27 sec\n"
     ]
    }
   ],
   "source": [
    "# Instanciate an Logistic Regression classifier\n",
    "clf = LogisticRegression(max_iter=1000)\n",
    "start_time = time.time()\n",
    "clf.fit(X_train_vectorized, y_train)\n",
    "end_time = time.time()\n",
    "elapsed_time = end_time - start_time\n",
    "\n",
    "# Get decision scores instead of binary predictions\n",
    "y_train_scores = clf.predict_proba(X_train_vectorized)\n",
    "y_validation_scores = clf.predict_proba(X_validation_vectorized)\n",
    "print(f\"Training time: {elapsed_time:.2f} sec\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8ca9c596-6516-4106-b48d-88fe234a44fe",
   "metadata": {},
   "source": [
    "### Grid search for optimal decision boundary threshold"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b366c81-7b44-4124-a38a-54d353f501d5",
   "metadata": {},
   "source": [
    "Now let's find the best decision boundary threshold for the positive and negative labels. This time, the threshold is bounded to a range between 0 and 1 because the Logistic regression outputs the probability of a sample belonging to a specific class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "5b76dd42-f19f-4025-af48-f2ae9476ac9b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Define the thresholds to test\n",
    "thresholds = np.arange(0, 1.01, 0.01)\n",
    "\n",
    "# Initialize empty lists to store accuracy and F1 scores\n",
    "train_accuracies = []\n",
    "validation_accuracies = []\n",
    "\n",
    "# Loop through each threshold\n",
    "for threshold in thresholds:\n",
    "    y_train_pred = np.where(y_train_scores[:, 1] >= threshold, 1, -1)\n",
    "    y_validation_pred = np.where(y_validation_scores[:, 1] >= threshold, 1, -1)\n",
    "\n",
    "    train_accuracy = accuracy_score(y_train, y_train_pred)\n",
    "    validation_accuracy = accuracy_score(y_validation, y_validation_pred)\n",
    "\n",
    "    train_accuracies.append(train_accuracy * 100)\n",
    "    validation_accuracies.append(validation_accuracy * 100)\n",
    "\n",
    "# Create a line plot for training set\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(thresholds, train_accuracies, label='Trainning set Accuracy')\n",
    "\n",
    "# Create a line plot for validation set\n",
    "plt.plot(thresholds, validation_accuracies, label='Validation set Accuracy')\n",
    "\n",
    "plt.xlabel('Threshold')\n",
    "plt.ylabel('Percentage')\n",
    "plt.title('Performance Metrics vs. Threshold')\n",
    "plt.legend()\n",
    "plt.grid(True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "83f4ef3a-2167-4f13-b783-5b757ae8be6e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The threshold for maximum validation accuracy 81.73% is: 0.53\n"
     ]
    }
   ],
   "source": [
    "# Find the index of the max validation accuracy\n",
    "max_accuracy_index = np.argmax(validation_accuracies)\n",
    "\n",
    "# Get the corresponding threshold\n",
    "optimal_threshold = thresholds[max_accuracy_index]\n",
    "\n",
    "print(\n",
    "    f\"The threshold for maximum validation accuracy {validation_accuracies[max_accuracy_index]:.2f}% is: {optimal_threshold:.2f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d5ee2347-c3be-427e-b65b-6d0019e337a0",
   "metadata": {},
   "source": [
    "Since the best threshold that we found is 0.53 (\t$\\approx$ 0.5), this is statistically coherent with the fact that the training data are balanced 50.2% negative and 49.8% positive. Therefore, we can keep the regular decision boundary of the Logistic Regression model which is 0.5."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91503512-1422-4869-b234-0c945a3f8941",
   "metadata": {},
   "source": [
    "### Grid search for optimal learning rate"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d67df6f6-23c0-4415-913b-637bb73c41a4",
   "metadata": {},
   "source": [
    "Logistic Regression involves two interconnected learning hyperparameters: `gamma` and `max_iter`. We will conduct a grid search to determine the best values for these parameters."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "7ce24987-3b22-41ee-900a-f5d82808292a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters: {'C': 1, 'max_iter': 500}\n",
      "Best score: 0.8155096953548681\n"
     ]
    }
   ],
   "source": [
    "# Define Logistic Regression classifier\n",
    "clf = LogisticRegression()\n",
    "\n",
    "# Define parameter grid for Logistic Regression\n",
    "param_grid = {\n",
    "    'C': [0.001, 0.01, 0.1, 1],\n",
    "    'max_iter': [500, 1000, 5000]\n",
    "}\n",
    "\n",
    "# Set up GridSearchCV\n",
    "grid_search = GridSearchCV(estimator=clf, param_grid=param_grid, n_jobs=-1, cv=5, scoring='accuracy')\n",
    "\n",
    "# Perform grid search\n",
    "grid_search.fit(X_train_vectorized, y_train)\n",
    "\n",
    "# After fitting, access the best parameters and score\n",
    "print(\"Best parameters:\", grid_search.best_params_)\n",
    "print(\"Best score:\", grid_search.best_score_)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1673719-ce27-465b-bb2b-fec8626f3ba0",
   "metadata": {},
   "source": [
    "### Grid search for optimal regularization"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f15cfcd0-b8a1-4459-8a0d-b790f1bcfcfa",
   "metadata": {},
   "source": [
    "A grid search was initially planned on three parameters :\n",
    "\n",
    "- `penalty`\n",
    "- `ratio`\n",
    "- `solver`\n",
    "\n",
    "However, it was discovered that the Logisitc Regression model does not support the combination of these parameters. Additionally, addressing solver incompatibilities led to issues with the gradient descent not converging. Given these challenges, the focus has been shifted to solely testing the l2 regularization, utilizing the model's default solver."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "855c5fb8-14d3-4e19-bfb9-c179d1f16aef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-2 {color: black;}#sk-container-id-2 pre{padding: 0;}#sk-container-id-2 div.sk-toggleable {background-color: white;}#sk-container-id-2 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-2 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-2 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-2 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-2 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-2 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-2 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-2 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-2 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-2 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-2 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-2 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-2 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-2 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-2 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-2 div.sk-item {position: relative;z-index: 1;}#sk-container-id-2 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-2 div.sk-item::before, #sk-container-id-2 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-2 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-2 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-2 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-2 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-2 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-2 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-2 div.sk-label-container {text-align: center;}#sk-container-id-2 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-2 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-2\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>GridSearchCV(cv=5, estimator=LogisticRegression(C=1, max_iter=500), n_jobs=-1,\n",
       "             param_grid={&#x27;penalty&#x27;: [&#x27;l2&#x27;]}, scoring=&#x27;accuracy&#x27;)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item sk-dashed-wrapped\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-4\" type=\"checkbox\" ><label for=\"sk-estimator-id-4\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">GridSearchCV</label><div class=\"sk-toggleable__content\"><pre>GridSearchCV(cv=5, estimator=LogisticRegression(C=1, max_iter=500), n_jobs=-1,\n",
       "             param_grid={&#x27;penalty&#x27;: [&#x27;l2&#x27;]}, scoring=&#x27;accuracy&#x27;)</pre></div></div></div><div class=\"sk-parallel\"><div class=\"sk-parallel-item\"><div class=\"sk-item\"><div class=\"sk-label-container\"><div class=\"sk-label sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-5\" type=\"checkbox\" ><label for=\"sk-estimator-id-5\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">estimator: LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression(C=1, max_iter=500)</pre></div></div></div><div class=\"sk-serial\"><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-6\" type=\"checkbox\" ><label for=\"sk-estimator-id-6\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression(C=1, max_iter=500)</pre></div></div></div></div></div></div></div></div></div></div>"
      ],
      "text/plain": [
       "GridSearchCV(cv=5, estimator=LogisticRegression(C=1, max_iter=500), n_jobs=-1,\n",
       "             param_grid={'penalty': ['l2']}, scoring='accuracy')"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define Logistic Regression classifier with optimal hyperparameters\n",
    "clf = LogisticRegression(C=1, max_iter=500)\n",
    "\n",
    "param_grid = {\n",
    "    'penalty': ['l2']\n",
    "}\n",
    "\n",
    "# param_grid = [\n",
    "#     {'penalty': ['l2'], 'solver': ['lbfgs', 'liblinear', 'saga']},\n",
    "#     {'penalty': ['l1'], 'solver': ['liblinear', 'saga']},\n",
    "#     {'penalty': ['elasticnet'], 'solver': ['saga'], 'l1_ratio': [0.001, 0.01, 0.1]}\n",
    "# ]\n",
    "\n",
    "# Set up GridSearchCV\n",
    "grid_search = GridSearchCV(estimator=clf, param_grid=param_grid, n_jobs=-1, cv=5, scoring='accuracy')\n",
    "\n",
    "# Perform grid search\n",
    "grid_search.fit(X_train_vectorized, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "99a168fa-31aa-4e49-a61c-6fa3a5e490a3",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best parameters: {'penalty': 'l2'}\n",
      "Best accuracy: 81.55%\n"
     ]
    }
   ],
   "source": [
    "# Get the best parameters\n",
    "best_parameters = grid_search.best_params_\n",
    "best_score = grid_search.best_score_\n",
    "\n",
    "print(\"Best parameters:\", best_parameters)\n",
    "print(f\"Best accuracy: {best_score * 100:.2f}%\")\n",
    "\n",
    "# Predictions with the best model\n",
    "y_pred = grid_search.predict(X_validation_vectorized)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3edb32eb-61bc-40d1-a849-fe5d0cdfe98f",
   "metadata": {},
   "source": [
    "We observe that the regularization doesn't affect the accuracy. From this, we can assume that our model is not overfitting the data."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "679bc8ee-ecfc-4868-b18c-6f6eb9b2ff17",
   "metadata": {
    "tags": []
   },
   "source": [
    "# 5. Submission using the Full Dataset"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f991f95b-48f9-4b75-8012-035ed473bc44",
   "metadata": {},
   "source": [
    "Now, let's proceed to test our optimal vectorizer and model on the full dataset (2.5M entries), and then make a submission on AIcrowd."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c24717ae-1a3a-4165-af44-fcf55cca6cd8",
   "metadata": {},
   "source": [
    "First, we load the full training dataset and the test data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "11253963-a178-45e6-8021-20329947620f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Load the full data\n",
    "full_data_train = load_preprocessed_full_data()\n",
    "full_data_test = load_preprocessed_test_data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "213084a1-5ea9-4039-97cb-f9e87e259dbf",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Prepare the data feature and target\n",
    "X_train_full = full_data_train[\"text\"]\n",
    "y_train_full = full_data_train[\"label\"]\n",
    "X_test = full_data_test[\"text\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c3c2b1db-e469-4dcb-b17e-e7739153ffdf",
   "metadata": {},
   "source": [
    "Then we setup the optimal vectorizer with the optimal model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "0a7d8ace-e29b-455a-8a7d-2471eab5800e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Use the optimal vectorizer: CountVectorizer with ngram_range=(1, 3)\n",
    "full_vectorizer = CountVectorizer(ngram_range=(1, 3))\n",
    "X_train_full_vectorized = full_vectorizer.fit_transform(X_train_full)\n",
    "X_test_vectorized = full_vectorizer.transform(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "9249f46d-e85f-4f20-b418-ba55f44b9403",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-3 {color: black;}#sk-container-id-3 pre{padding: 0;}#sk-container-id-3 div.sk-toggleable {background-color: white;}#sk-container-id-3 label.sk-toggleable__label {cursor: pointer;display: block;width: 100%;margin-bottom: 0;padding: 0.3em;box-sizing: border-box;text-align: center;}#sk-container-id-3 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-3 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-3 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-3 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-3 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-3 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-3 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-3 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 input.sk-hidden--visually {border: 0;clip: rect(1px 1px 1px 1px);clip: rect(1px, 1px, 1px, 1px);height: 1px;margin: -1px;overflow: hidden;padding: 0;position: absolute;width: 1px;}#sk-container-id-3 div.sk-estimator {font-family: monospace;background-color: #f0f8ff;border: 1px dotted black;border-radius: 0.25em;box-sizing: border-box;margin-bottom: 0.5em;}#sk-container-id-3 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-3 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-3 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-3 div.sk-serial::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: 0;}#sk-container-id-3 div.sk-serial {display: flex;flex-direction: column;align-items: center;background-color: white;padding-right: 0.2em;padding-left: 0.2em;position: relative;}#sk-container-id-3 div.sk-item {position: relative;z-index: 1;}#sk-container-id-3 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-3 div.sk-item::before, #sk-container-id-3 div.sk-parallel-item::before {content: \"\";position: absolute;border-left: 1px solid gray;box-sizing: border-box;top: 0;bottom: 0;left: 50%;z-index: -1;}#sk-container-id-3 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-3 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-3 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-3 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-3 div.sk-dashed-wrapped {border: 1px dashed gray;margin: 0 0.4em 0.5em 0.4em;box-sizing: border-box;padding-bottom: 0.4em;background-color: white;}#sk-container-id-3 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-3 div.sk-label-container {text-align: center;}#sk-container-id-3 div.sk-container {/* jupyter's `normalize.less` sets `[hidden] { display: none; }` but bootstrap.min.css set `[hidden] { display: none !important; }` so we also need the `!important` here to be able to override the default hidden behavior on the sphinx rendered scikit-learn.org. See: https://github.com/scikit-learn/scikit-learn/issues/21755 */display: inline-block !important;position: relative;}#sk-container-id-3 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-3\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>LogisticRegression(C=1, max_iter=500)</pre><b>In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook. <br />On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.</b></div><div class=\"sk-container\" hidden><div class=\"sk-item\"><div class=\"sk-estimator sk-toggleable\"><input class=\"sk-toggleable__control sk-hidden--visually\" id=\"sk-estimator-id-7\" type=\"checkbox\" checked><label for=\"sk-estimator-id-7\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">LogisticRegression</label><div class=\"sk-toggleable__content\"><pre>LogisticRegression(C=1, max_iter=500)</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "LogisticRegression(C=1, max_iter=500)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use the Logistic Regression classifier with optimal hyperparameters and without any regularization\n",
    "clf_submission = LogisticRegression(C=1, max_iter=500)\n",
    "clf.fit(X_train_full_vectorized, y_train_full)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ff634d3f-2783-437b-ac17-9c1992bff2b4",
   "metadata": {},
   "source": [
    "Finally, we can make the prediciton and make a submission."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "baea216d-c26b-436c-a527-5f226d181fb4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Make the prediction\n",
    "y_test_pred = clf.predict(X_test_vectorized)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "706f85ff-ced9-4860-b1f6-46e1a7982870",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Export the prediction\n",
    "export_submission(y_test_pred, f\"{DataPath.ML_SUBMISSION}/test_predictions_logistic_regression.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f7d5356-c7fd-4d3b-97d1-7b43013da1bf",
   "metadata": {},
   "source": [
    "This submission to AIcrowd achieved the following accuracy scores:\n",
    "    \n",
    "- First Score =`0.856`\n",
    "- Secondary Score = `0.859`\n",
    "\n",
    "You can access the results here:\n",
    "\n",
    "- csv output file : [test_predictions_logistic_regression.csv](../submissions/ml/test_predictions_logistic_regression.csv)\n",
    "- AIcrowd submission id : **[#247219](https://www.aicrowd.com/challenges/epfl-ml-text-classification/submissions/247219)**\n",
    "\n",
    "Although Logistic Regression requires more time compared to SGD, it has achieved an accuracy that is 2% higher. Let's explore a new approach : [Gated Recurrent Unit (GRU)](./model_GRU.ipynb) neural network."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "outputs": [],
   "source": [],
   "metadata": {
    "collapsed": false
   },
   "id": "ed7898be0182edc1"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:ada]",
   "language": "python",
   "name": "conda-env-ada-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}