{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2024-05-18T13:48:56.414451700Z",
     "start_time": "2024-05-18T13:48:54.900019200Z"
    }
   },
   "outputs": [],
   "source": [
    "# Read in the data and clean up column names\n",
    "import gensim\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "pd.set_option('display.max_colwidth', 100)\n",
    "data = pd.read_csv(\"train.tsv\", sep=\"\\t\", on_bad_lines='skip')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "outputs": [
    {
     "data": {
      "text/plain": "   label  \\\n0      1   \n1      0   \n2      1   \n3      0   \n4      1   \n\n                                                                                                  text  \n0  Przyjmujący reprezentacji Polski wrócił do PGE Skry Bełchatów Tylko rok trwał rozbrat Artura Sza...  \n1  FEN 9: Zapowiedź walki Róża Gumienna vs Katarzyna Posiadała (wideo) Podczas Fight Exclusive Nigh...  \n2  Aleksander Filipiak: Czuję się dobrze w nowym klubie Aleksander Filipiak w przerwie letniej zami...  \n3  Victoria Carl i Aleksiej Czerwotkin mistrzami świata juniorów na 5 i 10 kilometrów Biegi na 5 i ...  \n4  Świat poznał ją na mundialu. Francuska WAG czaruje pięknym ciałem Rachel Legrain-Trapani to jedn...  ",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>label</th>\n      <th>text</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>1</td>\n      <td>Przyjmujący reprezentacji Polski wrócił do PGE Skry Bełchatów Tylko rok trwał rozbrat Artura Sza...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>FEN 9: Zapowiedź walki Róża Gumienna vs Katarzyna Posiadała (wideo) Podczas Fight Exclusive Nigh...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>1</td>\n      <td>Aleksander Filipiak: Czuję się dobrze w nowym klubie Aleksander Filipiak w przerwie letniej zami...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>Victoria Carl i Aleksiej Czerwotkin mistrzami świata juniorów na 5 i 10 kilometrów Biegi na 5 i ...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>1</td>\n      <td>Świat poznał ją na mundialu. Francuska WAG czaruje pięknym ciałem Rachel Legrain-Trapani to jedn...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns = [\"label\", \"text\"]\n",
    "data.head()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T13:48:56.414451700Z",
     "start_time": "2024-05-18T13:48:56.401231900Z"
    }
   },
   "id": "9c4c96b33245043a"
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "outputs": [
    {
     "data": {
      "text/plain": "   label  \\\n0      1   \n1      0   \n2      1   \n3      0   \n4      1   \n\n                                                                                                  text  \\\n0  Przyjmujący reprezentacji Polski wrócił do PGE Skry Bełchatów Tylko rok trwał rozbrat Artura Sza...   \n1  FEN 9: Zapowiedź walki Róża Gumienna vs Katarzyna Posiadała (wideo) Podczas Fight Exclusive Nigh...   \n2  Aleksander Filipiak: Czuję się dobrze w nowym klubie Aleksander Filipiak w przerwie letniej zami...   \n3  Victoria Carl i Aleksiej Czerwotkin mistrzami świata juniorów na 5 i 10 kilometrów Biegi na 5 i ...   \n4  Świat poznał ją na mundialu. Francuska WAG czaruje pięknym ciałem Rachel Legrain-Trapani to jedn...   \n\n                                                                                            text_clean  \n0  [przyjmujący, reprezentacji, polski, wrócił, do, pge, skry, bełchatów, tylko, rok, trwał, rozbra...  \n1  [fen, zapowiedź, walki, róża, gumienna, vs, katarzyna, posiadała, wideo, podczas, fight, exclusi...  \n2  [aleksander, filipiak, czuję, się, dobrze, nowym, klubie, aleksander, filipiak, przerwie, letnie...  \n3  [victoria, carl, aleksiej, czerwotkin, mistrzami, świata, juniorów, na, kilometrów, biegi, na, k...  \n4  [świat, poznał, ją, na, mundialu, francuska, wag, czaruje, pięknym, ciałem, rachel, legrain, tra...  ",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>label</th>\n      <th>text</th>\n      <th>text_clean</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>1</td>\n      <td>Przyjmujący reprezentacji Polski wrócił do PGE Skry Bełchatów Tylko rok trwał rozbrat Artura Sza...</td>\n      <td>[przyjmujący, reprezentacji, polski, wrócił, do, pge, skry, bełchatów, tylko, rok, trwał, rozbra...</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>0</td>\n      <td>FEN 9: Zapowiedź walki Róża Gumienna vs Katarzyna Posiadała (wideo) Podczas Fight Exclusive Nigh...</td>\n      <td>[fen, zapowiedź, walki, róża, gumienna, vs, katarzyna, posiadała, wideo, podczas, fight, exclusi...</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>1</td>\n      <td>Aleksander Filipiak: Czuję się dobrze w nowym klubie Aleksander Filipiak w przerwie letniej zami...</td>\n      <td>[aleksander, filipiak, czuję, się, dobrze, nowym, klubie, aleksander, filipiak, przerwie, letnie...</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>0</td>\n      <td>Victoria Carl i Aleksiej Czerwotkin mistrzami świata juniorów na 5 i 10 kilometrów Biegi na 5 i ...</td>\n      <td>[victoria, carl, aleksiej, czerwotkin, mistrzami, świata, juniorów, na, kilometrów, biegi, na, k...</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>1</td>\n      <td>Świat poznał ją na mundialu. Francuska WAG czaruje pięknym ciałem Rachel Legrain-Trapani to jedn...</td>\n      <td>[świat, poznał, ją, na, mundialu, francuska, wag, czaruje, pięknym, ciałem, rachel, legrain, tra...</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['text_clean'] = data['text'].apply(lambda x: gensim.utils.simple_preprocess(x))\n",
    "data.head()"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T13:49:00.316282200Z",
     "start_time": "2024-05-18T13:48:56.410929500Z"
    }
   },
   "id": "e97f8679dd19876b"
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "outputs": [],
   "source": [
    "w2v_model = gensim.models.Word2Vec(data[\"text_clean\"],\n",
    "                                   vector_size=500,\n",
    "                                   window=5,\n",
    "                                   min_count=2,\n",
    "                                   workers=4)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:15:38.926464200Z",
     "start_time": "2024-05-18T14:15:22.986675300Z"
    }
   },
   "id": "c336ac485de74549"
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "outputs": [],
   "source": [
    "from keras.src.utils import pad_sequences\n",
    "from keras.src.legacy.preprocessing.text import Tokenizer\n",
    "\n",
    "token = Tokenizer(7229)\n",
    "token.fit_on_texts(data['text_clean'])\n",
    "text = token.texts_to_sequences(data['text_clean'])\n",
    "text = pad_sequences(text, 75)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:40:46.467152700Z",
     "start_time": "2024-05-18T14:40:43.515266200Z"
    }
   },
   "id": "738bc0b5f175c1d4"
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "outputs": [
    {
     "data": {
      "text/plain": "array([[   0,    0,    0, ...,    2,   20, 1957],\n       [   0,    0,    0, ..., 7153,  441,  292],\n       [   0,    0,    0, ..., 3702, 2385,    9],\n       ...,\n       [   0,    0,    0, ...,  520, 1094, 3132],\n       [   0,    0,    0, ...,   44,  287, 1800],\n       [   0,    0,    0, ...,  160,   57,  187]])"
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(np.array(text), data['label'], test_size=0.2)\n",
    "X_train"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T13:53:11.632921800Z",
     "start_time": "2024-05-18T13:53:11.609890400Z"
    }
   },
   "id": "8e68a73523198872"
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "outputs": [],
   "source": [
    "vocab_size = len(token.word_index) + 1"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T13:55:24.933067600Z",
     "start_time": "2024-05-18T13:55:24.892887400Z"
    }
   },
   "id": "ee4996706330dd07"
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "outputs": [],
   "source": [
    "embedding_matrix = np.zeros((vocab_size, 500))\n",
    "for word, i in token.word_index.items():\n",
    "    if word in w2v_model.wv:\n",
    "        embedding_matrix[i] = w2v_model.wv[word]"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:17:28.570268600Z",
     "start_time": "2024-05-18T14:17:28.377171400Z"
    }
   },
   "id": "80de5dc612e62c33"
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "outputs": [],
   "source": [
    "import keras\n",
    "\n",
    "opt = keras.optimizers.Adam(learning_rate=0.001)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:17:29.132463600Z",
     "start_time": "2024-05-18T14:17:29.117615100Z"
    }
   },
   "id": "fba513611f471de8"
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m30s\u001B[0m 22ms/step - acc: 0.9434 - loss: 0.1327 - val_acc: 0.9795 - val_loss: 0.0647\n",
      "Epoch 2/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m26s\u001B[0m 21ms/step - acc: 0.9801 - loss: 0.0614 - val_acc: 0.9798 - val_loss: 0.0675\n",
      "Epoch 3/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m27s\u001B[0m 22ms/step - acc: 0.9831 - loss: 0.0532 - val_acc: 0.9822 - val_loss: 0.0542\n",
      "Epoch 4/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m27s\u001B[0m 22ms/step - acc: 0.9851 - loss: 0.0455 - val_acc: 0.9832 - val_loss: 0.0502\n",
      "Epoch 5/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m27s\u001B[0m 22ms/step - acc: 0.9850 - loss: 0.0458 - val_acc: 0.9803 - val_loss: 0.0574\n",
      "Epoch 6/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m26s\u001B[0m 21ms/step - acc: 0.9860 - loss: 0.0419 - val_acc: 0.9836 - val_loss: 0.0552\n",
      "Epoch 7/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m27s\u001B[0m 22ms/step - acc: 0.9862 - loss: 0.0395 - val_acc: 0.9830 - val_loss: 0.0646\n",
      "Epoch 8/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m27s\u001B[0m 22ms/step - acc: 0.9874 - loss: 0.0362 - val_acc: 0.9787 - val_loss: 0.0723\n",
      "Epoch 9/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m34s\u001B[0m 27ms/step - acc: 0.9876 - loss: 0.0346 - val_acc: 0.9798 - val_loss: 0.0796\n",
      "Epoch 10/10\n",
      "\u001B[1m1227/1227\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m50s\u001B[0m 40ms/step - acc: 0.9885 - loss: 0.0347 - val_acc: 0.9829 - val_loss: 0.0487\n"
     ]
    },
    {
     "data": {
      "text/plain": "<keras.src.callbacks.history.History at 0x1ffe75a94d0>"
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from keras.src.layers import Dropout, Dense, Activation, Embedding, MaxPooling1D, GlobalMaxPooling1D\n",
    "from keras.src.layers import Conv1D\n",
    "from keras import Sequential\n",
    "\n",
    "keras_model = Sequential()\n",
    "keras_model.add(Embedding(vocab_size, 500, weights=[embedding_matrix], trainable=False))\n",
    "keras_model.add(Dropout(0.4))\n",
    "keras_model.add(Conv1D(50, 3, activation='relu', padding='same', strides=1))\n",
    "keras_model.add(Conv1D(50, 3, activation='relu', padding='same', strides=1))\n",
    "keras_model.add(MaxPooling1D())\n",
    "keras_model.add(Dropout(0.4))\n",
    "keras_model.add(Conv1D(100, 3, activation='relu', padding='same', strides=1))\n",
    "keras_model.add(Conv1D(100, 3, activation='relu', padding='same', strides=1))\n",
    "keras_model.add(MaxPooling1D())\n",
    "keras_model.add(Dropout(0.2))\n",
    "keras_model.add(Conv1D(200, 3, activation='relu', padding='same', strides=1))\n",
    "keras_model.add(Conv1D(200, 3, activation='relu', padding='same', strides=1))\n",
    "keras_model.add(GlobalMaxPooling1D())\n",
    "keras_model.add(Dropout(0.4))\n",
    "keras_model.add(Dense(200))\n",
    "keras_model.add(Activation('relu'))\n",
    "keras_model.add(Dropout(0.4))\n",
    "keras_model.add(Dense(1))\n",
    "keras_model.add(Activation('sigmoid'))\n",
    "keras_model.compile(loss='binary_crossentropy', metrics=['acc'], optimizer=opt)\n",
    "keras_model.fit(X_train, y_train, batch_size=64, epochs=10, validation_data=(X_test, y_test))"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:22:33.393859600Z",
     "start_time": "2024-05-18T14:17:33.183305300Z"
    }
   },
   "id": "81d4e85a23cdcc4d"
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "outputs": [],
   "source": [
    "model = keras_model"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:22:45.908737900Z",
     "start_time": "2024-05-18T14:22:45.901698900Z"
    }
   },
   "id": "8cac96bd584177b9"
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "outputs": [],
   "source": [
    "def preprocess(path):\n",
    "    data = pd.read_csv(path, sep=\"\\t\", on_bad_lines='skip')\n",
    "    data.columns = [\"text\"]\n",
    "    data['text_clean'] = data['text'].apply(lambda x: gensim.utils.simple_preprocess(x))\n",
    "    text = token.texts_to_sequences(data['text_clean'])\n",
    "    text = pad_sequences(text, 75)\n",
    "    return text"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:40:56.599000800Z",
     "start_time": "2024-05-18T14:40:56.587505200Z"
    }
   },
   "id": "104927eec2bcdcc8"
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "outputs": [],
   "source": [
    "x = preprocess(\"test-A/in.tsv\")"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:40:57.904309600Z",
     "start_time": "2024-05-18T14:40:57.622468400Z"
    }
   },
   "id": "4ac94af81d290ba5"
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001B[1m171/171\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m1s\u001B[0m 4ms/step\n"
     ]
    }
   ],
   "source": [
    "res = model.predict(x)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:40:59.833278700Z",
     "start_time": "2024-05-18T14:40:58.985503300Z"
    }
   },
   "id": "532a596a0cca6e3d"
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "outputs": [],
   "source": [
    "y_predictions = np.where(res>=0.49, 1, 0) "
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:41:01.471611600Z",
     "start_time": "2024-05-18T14:41:01.452033600Z"
    }
   },
   "id": "8688068e9edcd98c"
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "outputs": [],
   "source": [
    "out = pd.DataFrame(y_predictions)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:41:01.961092200Z",
     "start_time": "2024-05-18T14:41:01.952610400Z"
    }
   },
   "id": "e9df2cfccc9da22"
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "outputs": [
    {
     "data": {
      "text/plain": "      0\n0     1\n1     1\n2     0\n3     1\n4     1\n...  ..\n5439  1\n5440  1\n5441  1\n5442  0\n5443  1\n\n[5444 rows x 1 columns]",
      "text/html": "<div>\n<style scoped>\n    .dataframe tbody tr th:only-of-type {\n        vertical-align: middle;\n    }\n\n    .dataframe tbody tr th {\n        vertical-align: top;\n    }\n\n    .dataframe thead th {\n        text-align: right;\n    }\n</style>\n<table border=\"1\" class=\"dataframe\">\n  <thead>\n    <tr style=\"text-align: right;\">\n      <th></th>\n      <th>0</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>...</th>\n      <td>...</td>\n    </tr>\n    <tr>\n      <th>5439</th>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>5440</th>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>5441</th>\n      <td>1</td>\n    </tr>\n    <tr>\n      <th>5442</th>\n      <td>0</td>\n    </tr>\n    <tr>\n      <th>5443</th>\n      <td>1</td>\n    </tr>\n  </tbody>\n</table>\n<p>5444 rows × 1 columns</p>\n</div>"
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "out"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:41:03.483852600Z",
     "start_time": "2024-05-18T14:41:03.461120700Z"
    }
   },
   "id": "84595ca2b4d0c5ac"
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "outputs": [],
   "source": [
    "out.to_csv('out.tsv', sep=\"\\t\", index=False) "
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:41:14.478686200Z",
     "start_time": "2024-05-18T14:41:14.466508400Z"
    }
   },
   "id": "e94c7c3f0aeacf35"
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import pathlib\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib.ticker import (MultipleLocator, AutoMinorLocator)\n",
    "from sklearn.metrics import confusion_matrix as cm_sklearn\n",
    "from sklearn.metrics import precision_score\n",
    "from sklearn.metrics import recall_score\n",
    "from sklearn.metrics import f1_score\n",
    "\n",
    "def plot_discrimination_threshold(clf, X_test, y_test, argmax='f1', title='Metrics vs Discriminant Threshold', fig_size=(10, 8), dpi=100, save_fig_path=None):\n",
    "    \"\"\"\n",
    "    Plot precision, recall and f1-score vs discriminant threshold for the given pipeline model\n",
    "    Parameters\n",
    "    ----------\n",
    "    clf : estimator instance (either sklearn.Pipeline, imblearn.Pipeline or a classifier)\n",
    "        PRE-FITTED classifier or a PRE-FITTED Pipeline in which the last estimator is a classifier.\n",
    "    X_test : pandas.DataFrame of shape (n_samples, n_features)\n",
    "        Test features.\n",
    "    y_test : pandas.Series of shape (n_samples,)\n",
    "        Target values.\n",
    "    argmax : str, default: 'f1'\n",
    "        Annotate the threshold maximized by the supplied metric. Options: 'f1', 'precision', 'recall'\n",
    "    title : str, default ='FPR and FNR vs Discriminant Threshold'\n",
    "        Plot title.\n",
    "    fig_size : tuple, default = (10, 8)\n",
    "        Size (inches) of the plot.\n",
    "    dpi : int, default = 100\n",
    "        Image DPI.\n",
    "    save_fig_path : str, defaut=None\n",
    "        Full path where to save the plot. Will generate the folders if they don't exist already.\n",
    "    Returns\n",
    "    -------\n",
    "        fig : Matplotlib.pyplot.Figure\n",
    "            Figure from matplotlib\n",
    "        ax : Matplotlib.pyplot.Axe\n",
    "            Axe object from matplotlib\n",
    "    \"\"\"\n",
    "        \n",
    "    thresholds = np.linspace(0, 1, 100)\n",
    "    \n",
    "    precision_ls = []\n",
    "    recall_ls = []\n",
    "    f1_ls = []\n",
    "    fpr_ls = []\n",
    "    fnr_ls = []\n",
    "    \n",
    "    # obtain probabilities\n",
    "    probs = clf.predict(X_test)\n",
    "\n",
    "    for threshold in thresholds:   \n",
    "    \n",
    "        # obtain class prediction based on threshold\n",
    "        y_predictions = np.where(probs>=threshold, 1, 0) \n",
    "        \n",
    "        # obtain confusion matrix\n",
    "        tn, fp, fn, tp = cm_sklearn(y_test, y_predictions).ravel()\n",
    "        \n",
    "        # obtain FRP and FNR\n",
    "        FPR = fp / (tn + fp)\n",
    "        FNR = fn / (tp + fn)\n",
    "        \n",
    "        # obtain precision, recall and f1 scores\n",
    "        precision = precision_score(y_test, y_predictions, average='binary')\n",
    "        recall = recall_score(y_test, y_predictions, average='binary')\n",
    "        f1 = f1_score(y_test, y_predictions, average='binary')\n",
    "         \n",
    "        precision_ls.append(precision)\n",
    "        recall_ls.append(recall)\n",
    "        f1_ls.append(f1)\n",
    "        fpr_ls.append(FPR)\n",
    "        fnr_ls.append(FNR)\n",
    "              \n",
    "    metrics = pd.concat([\n",
    "        pd.Series(precision_ls),\n",
    "        pd.Series(recall_ls),\n",
    "        pd.Series(f1_ls),\n",
    "        pd.Series(fpr_ls),\n",
    "        pd.Series(fnr_ls)], axis=1)\n",
    "\n",
    "    metrics.columns = ['precision', 'recall', 'f1', 'fpr', 'fnr']\n",
    "    metrics.index = thresholds\n",
    "    \n",
    "    plt.rcParams[\"figure.facecolor\"] = 'white'\n",
    "    plt.rcParams[\"axes.facecolor\"] = 'white'\n",
    "    plt.rcParams[\"savefig.facecolor\"] = 'white'\n",
    "                \n",
    "    fig, ax = plt.subplots(1, 1, figsize=fig_size, dpi=dpi)\n",
    "    ax.plot(metrics['precision'], label='Precision')\n",
    "    ax.plot(metrics['recall'], label='Recall')\n",
    "    ax.plot(metrics['f1'], label='f1')\n",
    "    ax.plot(metrics['fpr'], label='False Positive Rate (FPR)', linestyle='dotted')\n",
    "    ax.plot(metrics['fnr'], label='False Negative Rate (FNR)', linestyle='dotted')\n",
    "    \n",
    "    # Draw a threshold line\n",
    "    disc_threshold = round(metrics[argmax].idxmax(), 2)\n",
    "    ax.axvline(x=metrics[argmax].idxmax(), color='black', linestyle='dashed', label=\"$t_r$=\"+str(disc_threshold))\n",
    "\n",
    "    ax.xaxis.set_major_locator(MultipleLocator(0.1))\n",
    "    ax.xaxis.set_major_formatter('{x:.1f}')\n",
    "    \n",
    "    ax.yaxis.set_major_locator(MultipleLocator(0.1))\n",
    "    ax.yaxis.set_major_formatter('{x:.1f}')\n",
    "\n",
    "    ax.xaxis.set_minor_locator(MultipleLocator(0.05))    \n",
    "    ax.yaxis.set_minor_locator(MultipleLocator(0.05))    \n",
    "\n",
    "    ax.tick_params(which='both', width=2)\n",
    "    ax.tick_params(which='major', length=7)\n",
    "    ax.tick_params(which='minor', length=4, color='black') \n",
    "    \n",
    "    plt.grid(True)\n",
    "    \n",
    "    plt.xlabel('Probability Threshold', fontsize=18)\n",
    "    plt.ylabel('Scores', fontsize=18)\n",
    "    plt.title(title, fontsize=18)\n",
    "    leg = ax.legend(loc='best', frameon=True, framealpha=0.7)\n",
    "    leg_frame = leg.get_frame()\n",
    "    leg_frame.set_color('gold')\n",
    "    plt.show()\n",
    "\n",
    "    if (save_fig_path != None):\n",
    "        path = pathlib.Path(save_fig_path)\n",
    "        path.parent.mkdir(parents=True, exist_ok=True)\n",
    "        fig.savefig(save_fig_path, dpi=dpi)\n",
    "\n",
    "    return fig, ax, disc_threshold"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:35:04.038699500Z",
     "start_time": "2024-05-18T14:35:04.033068300Z"
    }
   },
   "id": "f9d6a8e84ef28b4d"
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001B[1m614/614\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m2s\u001B[0m 4ms/step\n"
     ]
    },
    {
     "data": {
      "text/plain": "<Figure size 1000x800 with 1 Axes>",
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": "(<Figure size 1000x800 with 1 Axes>,\n <Axes: title={'center': 'Metrics vs Discriminant Threshold'}, xlabel='Probability Threshold', ylabel='Scores'>,\n 0.49)"
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "plot_discrimination_threshold(model,X_test, y_test)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:35:09.245565700Z",
     "start_time": "2024-05-18T14:35:04.261983600Z"
    }
   },
   "id": "f01f4db44a119b53"
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\u001B[1m171/171\u001B[0m \u001B[32m━━━━━━━━━━━━━━━━━━━━\u001B[0m\u001B[37m\u001B[0m \u001B[1m1s\u001B[0m 4ms/step\n"
     ]
    }
   ],
   "source": [
    "x = preprocess(\"dev-0/in.tsv\")\n",
    "res = model.predict(x)\n",
    "y_predictions = np.where(res >= 0.49, 1, 0)\n",
    "out = pd.DataFrame(y_predictions)\n",
    "out.to_csv('dev-0/out.tsv', sep=\"\\t\", index=False) "
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:50:05.273321200Z",
     "start_time": "2024-05-18T14:50:04.167585600Z"
    }
   },
   "id": "e4b7e4685d5422fa"
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "outputs": [],
   "source": [
    "from sklearn.metrics import accuracy_score\n",
    "y = pd.read_csv(\"./dev-0/expected.tsv\")\n",
    "score = accuracy_score(y_true=y, y_pred=out)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:53:19.968430700Z",
     "start_time": "2024-05-18T14:53:19.941257100Z"
    }
   },
   "id": "d8b05baeae6d0980"
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "outputs": [
    {
     "data": {
      "text/plain": "0.9814712896716199"
     },
     "execution_count": 87,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "score"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:55:05.443038600Z",
     "start_time": "2024-05-18T14:55:05.419651600Z"
    }
   },
   "id": "ed98ddb1ead2ae3"
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "outputs": [],
   "source": [
    "import math\n",
    "\n",
    "points = math.ceil(score * 7.0)"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:54:23.693419700Z",
     "start_time": "2024-05-18T14:54:23.684500400Z"
    }
   },
   "id": "8b90ccda303fdf83"
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "outputs": [
    {
     "data": {
      "text/plain": "7"
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "points"
   ],
   "metadata": {
    "collapsed": false,
    "ExecuteTime": {
     "end_time": "2024-05-18T14:54:23.815715800Z",
     "start_time": "2024-05-18T14:54:23.810173500Z"
    }
   },
   "id": "448ba41113b8218f"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}