014-PreTrained-Sentence-Toxic.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Pre-trained Vectors - Sentence -  Toxic Comments\n",
    "\n",
    "A corpus of manually labeled comments - classifying each comment by its type of toxicity is available on Kaggle. We will aim to do a binary classification of whether a comment is toxic or not\n",
    "\n",
    "This notebook uses **Pre-trained Sentence Embedding** from Spacy to do the task."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import keras\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "import vis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Get the Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Uncomment these shell lines to get the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "#!wget http://bit.do/deep_toxic_train -P data/\n",
    "#!mv data/deep_toxic_train data/train.zip"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv(\"data/train.zip\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "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>id</th>\n",
       "      <th>comment_text</th>\n",
       "      <th>toxic</th>\n",
       "      <th>severe_toxic</th>\n",
       "      <th>obscene</th>\n",
       "      <th>threat</th>\n",
       "      <th>insult</th>\n",
       "      <th>identity_hate</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0000997932d777bf</td>\n",
       "      <td>Explanation\\nWhy the edits made under my usern...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>000103f0d9cfb60f</td>\n",
       "      <td>D'aww! He matches this background colour I'm s...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>000113f07ec002fd</td>\n",
       "      <td>Hey man, I'm really not trying to edit war. It...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0001b41b1c6bb37e</td>\n",
       "      <td>\"\\nMore\\nI can't make any real suggestions on ...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0001d958c54c6e35</td>\n",
       "      <td>You, sir, are my hero. Any chance you remember...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 id                                       comment_text  toxic  \\\n",
       "0  0000997932d777bf  Explanation\\nWhy the edits made under my usern...      0   \n",
       "1  000103f0d9cfb60f  D'aww! He matches this background colour I'm s...      0   \n",
       "2  000113f07ec002fd  Hey man, I'm really not trying to edit war. It...      0   \n",
       "3  0001b41b1c6bb37e  \"\\nMore\\nI can't make any real suggestions on ...      0   \n",
       "4  0001d958c54c6e35  You, sir, are my hero. Any chance you remember...      0   \n",
       "\n",
       "   severe_toxic  obscene  threat  insult  identity_hate  \n",
       "0             0        0       0       0              0  \n",
       "1             0        0       0       0              0  \n",
       "2             0        0       0       0              0  \n",
       "3             0        0       0       0              0  \n",
       "4             0        0       0       0              0  "
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Create the Input & Output Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    Explanation\\nWhy the edits made under my usern...\n",
       "1    D'aww! He matches this background colour I'm s...\n",
       "2    Hey man, I'm really not trying to edit war. It...\n",
       "3    \"\\nMore\\nI can't make any real suggestions on ...\n",
       "4    You, sir, are my hero. Any chance you remember...\n",
       "Name: comment_text, dtype: object"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_sentences = df[\"comment_text\"]\n",
    "train_sentences.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = df.iloc[:,2].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.utils import to_categorical\n",
    "y = to_categorical(labels)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Creating the sentence embedding for learning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "import spacy\n",
    "#!python -m spacy download en_core_web_md\n",
    "nlp = spacy.load('en_core_web_md')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<svg xmlns=\"http://www.w3.org/2000/svg\" xmlns:xlink=\"http://www.w3.org/1999/xlink\" id=\"0\" class=\"displacy\" width=\"750\" height=\"312.0\" style=\"max-width: none; height: 312.0px; color: #000000; background: #ffffff; font-family: Arial\">\n",
       "<text class=\"displacy-token\" fill=\"currentColor\" text-anchor=\"middle\" y=\"222.0\">\n",
       "    <tspan class=\"displacy-word\" fill=\"currentColor\" x=\"50\">This</tspan>\n",
       "    <tspan class=\"displacy-tag\" dy=\"2em\" fill=\"currentColor\" x=\"50\">DET</tspan>\n",
       "</text>\n",
       "\n",
       "<text class=\"displacy-token\" fill=\"currentColor\" text-anchor=\"middle\" y=\"222.0\">\n",
       "    <tspan class=\"displacy-word\" fill=\"currentColor\" x=\"225\">is</tspan>\n",
       "    <tspan class=\"displacy-tag\" dy=\"2em\" fill=\"currentColor\" x=\"225\">VERB</tspan>\n",
       "</text>\n",
       "\n",
       "<text class=\"displacy-token\" fill=\"currentColor\" text-anchor=\"middle\" y=\"222.0\">\n",
       "    <tspan class=\"displacy-word\" fill=\"currentColor\" x=\"400\">some</tspan>\n",
       "    <tspan class=\"displacy-tag\" dy=\"2em\" fill=\"currentColor\" x=\"400\">DET</tspan>\n",
       "</text>\n",
       "\n",
       "<text class=\"displacy-token\" fill=\"currentColor\" text-anchor=\"middle\" y=\"222.0\">\n",
       "    <tspan class=\"displacy-word\" fill=\"currentColor\" x=\"575\">text</tspan>\n",
       "    <tspan class=\"displacy-tag\" dy=\"2em\" fill=\"currentColor\" x=\"575\">NOUN</tspan>\n",
       "</text>\n",
       "\n",
       "<g class=\"displacy-arrow\">\n",
       "    <path class=\"displacy-arc\" id=\"arrow-0-0\" stroke-width=\"2px\" d=\"M70,177.0 C70,89.5 220.0,89.5 220.0,177.0\" fill=\"none\" stroke=\"currentColor\"/>\n",
       "    <text dy=\"1.25em\" style=\"font-size: 0.8em; letter-spacing: 1px\">\n",
       "        <textPath xlink:href=\"#arrow-0-0\" class=\"displacy-label\" startOffset=\"50%\" fill=\"currentColor\" text-anchor=\"middle\">nsubj</textPath>\n",
       "    </text>\n",
       "    <path class=\"displacy-arrowhead\" d=\"M70,179.0 L62,167.0 78,167.0\" fill=\"currentColor\"/>\n",
       "</g>\n",
       "\n",
       "<g class=\"displacy-arrow\">\n",
       "    <path class=\"displacy-arc\" id=\"arrow-0-1\" stroke-width=\"2px\" d=\"M420,177.0 C420,89.5 570.0,89.5 570.0,177.0\" fill=\"none\" stroke=\"currentColor\"/>\n",
       "    <text dy=\"1.25em\" style=\"font-size: 0.8em; letter-spacing: 1px\">\n",
       "        <textPath xlink:href=\"#arrow-0-1\" class=\"displacy-label\" startOffset=\"50%\" fill=\"currentColor\" text-anchor=\"middle\">det</textPath>\n",
       "    </text>\n",
       "    <path class=\"displacy-arrowhead\" d=\"M420,179.0 L412,167.0 428,167.0\" fill=\"currentColor\"/>\n",
       "</g>\n",
       "\n",
       "<g class=\"displacy-arrow\">\n",
       "    <path class=\"displacy-arc\" id=\"arrow-0-2\" stroke-width=\"2px\" d=\"M245,177.0 C245,2.0 575.0,2.0 575.0,177.0\" fill=\"none\" stroke=\"currentColor\"/>\n",
       "    <text dy=\"1.25em\" style=\"font-size: 0.8em; letter-spacing: 1px\">\n",
       "        <textPath xlink:href=\"#arrow-0-2\" class=\"displacy-label\" startOffset=\"50%\" fill=\"currentColor\" text-anchor=\"middle\">attr</textPath>\n",
       "    </text>\n",
       "    <path class=\"displacy-arrowhead\" d=\"M575.0,179.0 L583.0,167.0 567.0,167.0\" fill=\"currentColor\"/>\n",
       "</g>\n",
       "</svg>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "doc = nlp(\"This is some text\")\n",
    "from spacy import displacy\n",
    "displacy.render(doc, jupyter=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(300,)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Word Vector\n",
    "doc[3].vector.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(300,)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Sentence Vector\n",
    "doc.vector.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_vector(sentence):\n",
    "    vector = nlp(sentence).vector\n",
    "    return vector"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "from tqdm import tqdm, tqdm_pandas\n",
    "tqdm.pandas(desc=\"progress\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "progress: 100%|██████████| 159571/159571 [47:30<00:00, 55.99it/s]\n"
     ]
    }
   ],
   "source": [
    "X = train_sentences.progress_apply(get_vector)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "Xvec = np.vstack(X)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(159571, 300)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Xvec.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((127656, 300), (31915, 300), (127656, 2), (31915, 2))"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(Xvec, y, test_size=0.2, random_state=42)\n",
    "X_train.shape, X_test.shape, y_train.shape, y_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 2: Create the Model Architecture"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = Sequential()\n",
    "model.add(Dense(128, input_dim = 300, activation = \"relu\"))\n",
    "model.add(Dropout(0.25))\n",
    "model.add(Dense(64, activation = \"relu\"))\n",
    "model.add(Dropout(0.25))\n",
    "model.add(Dense(2, activation=\"sigmoid\"))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 3: Compile the Model & Fit on the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "model.compile(loss='binary_crossentropy', optimizer='rmsprop', metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train on 102124 samples, validate on 25532 samples\n",
      "Epoch 1/5\n",
      "102124/102124 [==============================] - 4s 38us/step - loss: 0.1519 - acc: 0.9436 - val_loss: 0.1470 - val_acc: 0.9392\n",
      "Epoch 2/5\n",
      "102124/102124 [==============================] - 3s 29us/step - loss: 0.1249 - acc: 0.9531 - val_loss: 0.1228 - val_acc: 0.9541\n",
      "Epoch 3/5\n",
      "102124/102124 [==============================] - 3s 29us/step - loss: 0.1219 - acc: 0.9545 - val_loss: 0.1210 - val_acc: 0.9554\n",
      "Epoch 4/5\n",
      "102124/102124 [==============================] - 3s 29us/step - loss: 0.1181 - acc: 0.9561 - val_loss: 0.1193 - val_acc: 0.9547\n",
      "Epoch 5/5\n",
      "102124/102124 [==============================] - 3s 29us/step - loss: 0.1163 - acc: 0.9568 - val_loss: 0.1241 - val_acc: 0.9549\n"
     ]
    }
   ],
   "source": [
    "output = model.fit(X_train, y_train, batch_size=128, epochs=5, validation_split=0.2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 4: Evaluate the Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.vegalite.v2+json": {
       "$schema": "https://vega.github.io/schema/vega-lite/v2.4.3.json",
       "config": {
        "view": {
         "height": 300,
         "width": 400
        }
       },
       "data": {
        "values": [
         {
          "batch": 1,
          "metrics": "loss",
          "type": "validation",
          "value": 0.147026669772725
         },
         {
          "batch": 2,
          "metrics": "loss",
          "type": "validation",
          "value": 0.12277916904416891
         },
         {
          "batch": 3,
          "metrics": "loss",
          "type": "validation",
          "value": 0.12102726382950005
         },
         {
          "batch": 4,
          "metrics": "loss",
          "type": "validation",
          "value": 0.11930778381239186
         },
         {
          "batch": 5,
          "metrics": "loss",
          "type": "validation",
          "value": 0.12405743738821676
         },
         {
          "batch": 1,
          "metrics": "acc",
          "type": "validation",
          "value": 0.9392135360109083
         },
         {
          "batch": 2,
          "metrics": "acc",
          "type": "validation",
          "value": 0.9540968197239579
         },
         {
          "batch": 3,
          "metrics": "acc",
          "type": "validation",
          "value": 0.9553697320448649
         },
         {
          "batch": 4,
          "metrics": "acc",
          "type": "validation",
          "value": 0.9547430675697984
         },
         {
          "batch": 5,
          "metrics": "acc",
          "type": "validation",
          "value": 0.9548605670754147
         },
         {
          "batch": 1,
          "metrics": "loss",
          "type": "training",
          "value": 0.15185227187281886
         },
         {
          "batch": 2,
          "metrics": "loss",
          "type": "training",
          "value": 0.12485747819599405
         },
         {
          "batch": 3,
          "metrics": "loss",
          "type": "training",
          "value": 0.12185092075552019
         },
         {
          "batch": 4,
          "metrics": "loss",
          "type": "training",
          "value": 0.11810195936075898
         },
         {
          "batch": 5,
          "metrics": "loss",
          "type": "training",
          "value": 0.11628440463795309
         },
         {
          "batch": 1,
          "metrics": "acc",
          "type": "training",
          "value": 0.9436420430345562
         },
         {
          "batch": 2,
          "metrics": "acc",
          "type": "training",
          "value": 0.9530570679016116
         },
         {
          "batch": 3,
          "metrics": "acc",
          "type": "training",
          "value": 0.9544964944670564
         },
         {
          "batch": 4,
          "metrics": "acc",
          "type": "training",
          "value": 0.9561023853588205
         },
         {
          "batch": 5,
          "metrics": "acc",
          "type": "training",
          "value": 0.9567878265871312
         }
        ]
       },
       "facet": {
        "column": {
         "field": "metrics",
         "type": "nominal"
        }
       },
       "resolve": {
        "scale": {
         "color": "shared",
         "x": "shared",
         "y": "independent"
        }
       },
       "spec": {
        "layer": [
         {
          "encoding": {
           "color": {
            "field": "type",
            "type": "nominal"
           },
           "tooltip": [
            {
             "field": "batch",
             "type": "quantitative"
            },
            {
             "field": "value",
             "type": "quantitative"
            }
           ],
           "x": {
            "field": "batch",
            "type": "quantitative"
           },
           "y": {
            "field": "value",
            "type": "quantitative"
           }
          },
          "height": 300,
          "mark": {
           "size": 50,
           "type": "circle"
          },
          "width": 300
         },
         {
          "encoding": {
           "color": {
            "field": "type",
            "type": "nominal"
           },
           "x": {
            "field": "batch",
            "type": "quantitative"
           },
           "y": {
            "field": "value",
            "type": "quantitative"
           }
          },
          "height": 300,
          "mark": "line",
          "width": 300
         }
        ]
       }
      },
      "image/png": "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",
      "text/plain": [
       "<VegaLite 2 object>\n",
       "\n",
       "If you see this message, it means the renderer has not been properly enabled\n",
       "for the frontend that you are using. For more information, see\n",
       "https://altair-viz.github.io/user_guide/troubleshooting.html\n"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "vis.metrics(output.history)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "31915/31915 [==============================] - 1s 33us/step\n"
     ]
    }
   ],
   "source": [
    "score = model.evaluate(X_test, y_test, verbose=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test loss: 0.12506679729801462\n",
      "Test accuracy: 0.9540811530628232\n"
     ]
    }
   ],
   "source": [
    "print('Test loss:', score[0])\n",
    "print('Test accuracy:', score[1])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Step 5: Visualise evaluation & Make a prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "predict_classes = model.predict_classes(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "actual_classes = np.dot(y_test,np.array([[0],[1]])).reshape(-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "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>col_0</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>row_0</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.0</th>\n",
       "      <td>28600</td>\n",
       "      <td>259</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.0</th>\n",
       "      <td>1209</td>\n",
       "      <td>1847</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "col_0      0     1\n",
       "row_0             \n",
       "0.0    28600   259\n",
       "1.0     1209  1847"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(actual_classes, predict_classes)"
   ]
  },
  {
   "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": "3.6.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}