013 NLP - Simple Sentiment Analysis using Bayes and Logistic Regression.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step #1 Load the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "scrolled": true
   },
   "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>textID</th>\n",
       "      <th>text</th>\n",
       "      <th>selected_text</th>\n",
       "      <th>sentiment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>cb774db0d1</td>\n",
       "      <td>I`d have responded, if I were going</td>\n",
       "      <td>I`d have responded, if I were going</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>549e992a42</td>\n",
       "      <td>Sooo SAD I will miss you here in San Diego!!!</td>\n",
       "      <td>Sooo SAD</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>088c60f138</td>\n",
       "      <td>my boss is bullying me...</td>\n",
       "      <td>bullying me</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>9642c003ef</td>\n",
       "      <td>what interview! leave me alone</td>\n",
       "      <td>leave me alone</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>358bd9e861</td>\n",
       "      <td>Sons of ****, why couldn`t they put them on t...</td>\n",
       "      <td>Sons of ****,</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27476</th>\n",
       "      <td>4eac33d1c0</td>\n",
       "      <td>wish we could come see u on Denver  husband l...</td>\n",
       "      <td>d lost</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27477</th>\n",
       "      <td>4f4c4fc327</td>\n",
       "      <td>I`ve wondered about rake to.  The client has ...</td>\n",
       "      <td>, don`t force</td>\n",
       "      <td>negative</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27478</th>\n",
       "      <td>f67aae2310</td>\n",
       "      <td>Yay good for both of you. Enjoy the break - y...</td>\n",
       "      <td>Yay good for both of you.</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27479</th>\n",
       "      <td>ed167662a5</td>\n",
       "      <td>But it was worth it  ****.</td>\n",
       "      <td>But it was worth it  ****.</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27480</th>\n",
       "      <td>6f7127d9d7</td>\n",
       "      <td>All this flirting going on - The ATG smiles...</td>\n",
       "      <td>All this flirting going on - The ATG smiles. Y...</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>27481 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           textID                                               text  \\\n",
       "0      cb774db0d1                I`d have responded, if I were going   \n",
       "1      549e992a42      Sooo SAD I will miss you here in San Diego!!!   \n",
       "2      088c60f138                          my boss is bullying me...   \n",
       "3      9642c003ef                     what interview! leave me alone   \n",
       "4      358bd9e861   Sons of ****, why couldn`t they put them on t...   \n",
       "...           ...                                                ...   \n",
       "27476  4eac33d1c0   wish we could come see u on Denver  husband l...   \n",
       "27477  4f4c4fc327   I`ve wondered about rake to.  The client has ...   \n",
       "27478  f67aae2310   Yay good for both of you. Enjoy the break - y...   \n",
       "27479  ed167662a5                         But it was worth it  ****.   \n",
       "27480  6f7127d9d7     All this flirting going on - The ATG smiles...   \n",
       "\n",
       "                                           selected_text sentiment  \n",
       "0                    I`d have responded, if I were going   neutral  \n",
       "1                                               Sooo SAD  negative  \n",
       "2                                            bullying me  negative  \n",
       "3                                         leave me alone  negative  \n",
       "4                                          Sons of ****,  negative  \n",
       "...                                                  ...       ...  \n",
       "27476                                             d lost  negative  \n",
       "27477                                      , don`t force  negative  \n",
       "27478                          Yay good for both of you.  positive  \n",
       "27479                         But it was worth it  ****.  positive  \n",
       "27480  All this flirting going on - The ATG smiles. Y...   neutral  \n",
       "\n",
       "[27481 rows x 4 columns]"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# A tutorial for this file is available at www.relataly.com\n",
    "\n",
    "import pandas as pd \n",
    "import matplotlib.pyplot as plt \n",
    "from sklearn.pipeline import Pipeline\n",
    "from sklearn.feature_extraction.text import CountVectorizer, TfidfTransformer\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn.naive_bayes import MultinomialNB\n",
    "from sklearn.metrics import classification_report, confusion_matrix, precision_recall_fscore_support as score\n",
    "import seaborn as sns\n",
    "sns.set_style('white', { 'axes.spines.right': False, 'axes.spines.top': False})\n",
    "\n",
    "# Load the train data\n",
    "path = 'data/sentiment/'\n",
    "train_base_df = pd.read_csv(path + \"train.csv\") \n",
    "test_base_df = pd.read_csv(path + \"test.csv\") \n",
    "\n",
    "# Print a Summary of the Train Data\n",
    "# The train data comprises 27481 text phrases \n",
    "# Sentiment Classes are labeled either \"neutral\", \"negative\" or \"positive\"\n",
    "train_base_df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step #2 Clean and Preprocess the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>textID</th>\n",
       "      <th>text</th>\n",
       "      <th>selected_text</th>\n",
       "      <th>sentiment</th>\n",
       "      <th>text_length</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>14339</th>\n",
       "      <td>5c6abc28a1</td>\n",
       "      <td>ow</td>\n",
       "      <td>ow</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26005</th>\n",
       "      <td>0b3fe0ca78</td>\n",
       "      <td>?</td>\n",
       "      <td>?</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11524</th>\n",
       "      <td>4105b6a05d</td>\n",
       "      <td>aw</td>\n",
       "      <td>aw</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>641</th>\n",
       "      <td>5210cc55ae</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25699</th>\n",
       "      <td>ee8ee67cb3</td>\n",
       "      <td>ME</td>\n",
       "      <td>ME</td>\n",
       "      <td>2</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2582</th>\n",
       "      <td>72df0ca5a2</td>\n",
       "      <td>TODAy SUCKs - kisskass19: ï¿½you and katey bro...</td>\n",
       "      <td>SUCKs</td>\n",
       "      <td>1</td>\n",
       "      <td>139.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16851</th>\n",
       "      <td>a7595518c2</td>\n",
       "      <td>4 novos followers http://migre.me/SfN Sorry, t...</td>\n",
       "      <td>Sorry,</td>\n",
       "      <td>1</td>\n",
       "      <td>139.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13896</th>\n",
       "      <td>3312d3de45</td>\n",
       "      <td>Feeling madly sorry for myself... - ï¿½Iï¿½M S...</td>\n",
       "      <td>madly sorry</td>\n",
       "      <td>1</td>\n",
       "      <td>140.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27318</th>\n",
       "      <td>d370238b6b</td>\n",
       "      <td>just saw an advert for ATTICS TO EDEN on tv  o...</td>\n",
       "      <td>just saw an advert for ATTICS TO EDEN on tv  o...</td>\n",
       "      <td>2</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3138</th>\n",
       "      <td>0d64ba9efd</td>\n",
       "      <td>Is so freaking bored on the bus. Hate being po...</td>\n",
       "      <td>bored</td>\n",
       "      <td>1</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>27480 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           textID                                               text  \\\n",
       "14339  5c6abc28a1                                                 ow   \n",
       "26005  0b3fe0ca78                                                  ?   \n",
       "11524  4105b6a05d                                                 aw   \n",
       "641    5210cc55ae                                                 no   \n",
       "25699  ee8ee67cb3                                                 ME   \n",
       "...           ...                                                ...   \n",
       "2582   72df0ca5a2  TODAy SUCKs - kisskass19: ï¿½you and katey bro...   \n",
       "16851  a7595518c2  4 novos followers http://migre.me/SfN Sorry, t...   \n",
       "13896  3312d3de45  Feeling madly sorry for myself... - ï¿½Iï¿½M S...   \n",
       "27318  d370238b6b  just saw an advert for ATTICS TO EDEN on tv  o...   \n",
       "3138   0d64ba9efd  Is so freaking bored on the bus. Hate being po...   \n",
       "\n",
       "                                           selected_text  sentiment  \\\n",
       "14339                                                 ow          2   \n",
       "26005                                                  ?          2   \n",
       "11524                                                 aw          2   \n",
       "641                                                   no          2   \n",
       "25699                                                 ME          2   \n",
       "...                                                  ...        ...   \n",
       "2582                                               SUCKs          1   \n",
       "16851                                             Sorry,          1   \n",
       "13896                                        madly sorry          1   \n",
       "27318  just saw an advert for ATTICS TO EDEN on tv  o...          2   \n",
       "3138                                               bored          1   \n",
       "\n",
       "       text_length  \n",
       "14339          3.0  \n",
       "26005          3.0  \n",
       "11524          3.0  \n",
       "641            3.0  \n",
       "25699          3.0  \n",
       "...            ...  \n",
       "2582         139.0  \n",
       "16851        139.0  \n",
       "13896        140.0  \n",
       "27318        141.0  \n",
       "3138         141.0  \n",
       "\n",
       "[27480 rows x 5 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define Class Integer Values\n",
    "cleanup_nums = {\"sentiment\": {\"negative\": 1, \"neutral\": 2, \"positive\": 3}}\n",
    "\n",
    "# Replace the Classes with Integer Values\n",
    "train_df = train_base_df.copy()\n",
    "train_df.replace(cleanup_nums, inplace=True)\n",
    "\n",
    "# Clean the Test Data\n",
    "test_df = test_base_df.copy()\n",
    "test_df.replace(cleanup_nums, inplace=True)\n",
    "\n",
    "# Create a Feature based on Text Length\n",
    "train_df['text_length'] = train_df['text'].str.len() # Store string length of each sample\n",
    "train_df = train_df.sort_values(['text_length'], ascending=True)\n",
    "train_df = train_df.dropna()\n",
    "train_df "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step #3 Explore the Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Labels')"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Print the Distribution of Sentiment Labels\n",
    "ax = sns.histplot(train_df, x='sentiment', bins=3, hue='sentiment', palette=['#e74c3c', \"#3498db\", '#2ecc71'])\n",
    "ax.set_xlabel('Count')\n",
    "ax.set_ylabel('Labels')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Text Length Distribution')"
      ]
     },
     "execution_count": 97,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize a distribution of text_length\n",
    "fig, ax = plt.subplots(figsize=(12,6))\n",
    "sns.histplot(data=train_df, x='text_length', bins='auto', hue='sentiment', palette=['#e74c3c', \"#3498db\", '#2ecc71'], alpha=.5, kde=True);\n",
    "plt.title('Text Length Distribution')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step #4 Train a Logistic Regression Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score      support\n",
      "negative       0.736479  0.639361  0.684492  1001.000000\n",
      "neutral        0.643070  0.755944  0.694953  1430.000000\n",
      "positive       0.795732  0.709882  0.750359  1103.000000\n",
      "accuracy       0.708546  0.708546  0.708546     0.708546\n",
      "macro avg      0.725093  0.701729  0.709935  3534.000000\n",
      "weighted avg   0.717175  0.708546  0.709283  3534.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create a transformation pipeline\n",
    "# The pipeline sequentially applies a list of transforms and as a final estimator logistic regression \n",
    "pipeline_log = Pipeline([\n",
    "                ('count', CountVectorizer()),\n",
    "                ('tfidf', TfidfTransformer()),\n",
    "                ('clf', LogisticRegression(solver='liblinear', multi_class='auto')),\n",
    "        ])\n",
    "\n",
    "# Train model using the created sklearn pipeline\n",
    "model_name = 'logistic regression classifier'\n",
    "model_lgr = pipeline_log.fit(train_df['text'], train_df['sentiment'])\n",
    "\n",
    "def evaluate_results(model, test_df):\n",
    "    # Predict class labels using the learner function\n",
    "    test_df['pred'] = model.predict(test_df['text'])\n",
    "    y_true = test_df['sentiment']\n",
    "    y_pred = test_df['pred']\n",
    "    target_names = ['negative', 'neutral', 'positive']\n",
    "\n",
    "    # Print the Confusion Matrix\n",
    "    results_log = classification_report(y_true, y_pred, target_names=target_names, output_dict=True)\n",
    "    results_df_log = pd.DataFrame(results_log).transpose()\n",
    "    print(results_df_log)\n",
    "    matrix = confusion_matrix(y_true,  y_pred)\n",
    "    sns.heatmap(pd.DataFrame(matrix), \n",
    "                annot=True, fmt=\"d\", linewidths=.5, cmap=\"YlGnBu\")\n",
    "    plt.xlabel('Predictions')\n",
    "    plt.xlabel('Actual')\n",
    "    \n",
    "    model_score = score(y_pred, y_true, average='macro')\n",
    "    return model_score\n",
    "\n",
    "    \n",
    "# Evaluate model performance\n",
    "model_score = evaluate_results(model_lgr, test_df)\n",
    "performance_df = pd.DataFrame().append({'model_name': model_name, \n",
    "                                    'f1_score': model_score[0], \n",
    "                                    'precision': model_score[1], \n",
    "                                    'recall': model_score[2]}, ignore_index=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step #4b Train a Bayes Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score      support\n",
      "negative       0.782700  0.370629  0.503051  1001.000000\n",
      "neutral        0.531678  0.868531  0.659586  1430.000000\n",
      "positive       0.821823  0.539438  0.651341  1103.000000\n",
      "accuracy       0.624788  0.624788  0.624788     0.624788\n",
      "macro avg      0.712067  0.592866  0.604659  3534.000000\n",
      "weighted avg   0.693337  0.624788  0.612674  3534.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWMAAAEECAYAAAD9M1UDAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAiMUlEQVR4nO3de3RU5b3G8e9cMrnMJEREK4IBAwSkNnKJoRTBQzFGPYcqFAgZjFrFVg4KiRa5BILiBRWMFzQiqMc2XBIs2gNVWwWVINKURgFRUYkVBBGIBJgZYDLJzPkDOxVPkSQm7D3D81lrr5U9s7P3b5btwy/vft89llAoFEJERAxlNboAERFRGIuImILCWETEBBTGIiImoDAWETEBu9EFiIi0lPiU3EYfe2TH0laspOlaPYxX7ni1tS9xWhuaciVDX19rdBlRbWXWwG9++sTQOqJbmtEFGE6dsYhEDYslckdeFcYiEjWslsiNtMitXETkOyK5M47cykVEvsNisTR6a4xNmzaRl5cHwEcffYTb7SYvL4+bbrqJmpoaAJYtW8bw4cMZNWoUb775JgBHjx7ltttuw+12c/PNN7N///6TXkthLCJRxNqE7fstXLiQ6dOn4/f7AbjvvvuYMWMGpaWlZGVlsXDhQvbt20dpaSllZWU8++yzFBcXU1dXx9KlS0lLS2PJkiVcc801lJSUNKpyEZGoYLFYG72dTEpKCvPmzQvvFxcXc8EFFwDQ0NBAbGwsmzdvpnfv3jgcDhITE0lJSWHr1q1UVVUxcOCxWTiDBg1i/fr1J72exoxFJGo0Zcy4vLyc8vLy8H5OTg45OTnh/ezsbHbu3BneP/vsswF49913WbRoEYsXL2bt2rUkJiaGj3E6nXi9Xrxeb/h1p9OJx+M5aT0KYxGJGk2ZTfHd8G2MV155haeeeooFCxbQtm1bXC4XPp8v/L7P5yMxMfG4130+H0lJSSevvUmViIiYWEsOU3zX//7v/7Jo0SJKS0s577zzAEhPT6eqqgq/34/H46G6upq0tDT69OnDmjVrAKioqKBv374nPb86YxGJGq01ta2hoYH77ruP9u3bc9tttwFw8cUXM2HCBPLy8nC73YRCIQoKCoiNjSU3N5fJkyeTm5tLTEwMDz/88Mlrb+1v+tBy6Nal5dCtT8uhT4WWWQ59VveCRh+77+NHWuSaLUWdsYhEjUhe9KEwFpGoYbVGbqRFbuUiIv+POmMREcNpmEJExAQUxiIiJmDRMIWIiPHUGYuImIDVajO6hGZTGItI1NAwhYiICWiYQkTEBBTGIiImoGEKERETsGg5tIiI8Rr7RaNmpDAWkaihYQoRERPQDTwRETPQMIWIiAlEbmOsMBaRKGKN3DRWGItI9IjcLD79wjjYEOSFR8rY98U+LDYLOb918+f/eZlD+w8BULtnP50u6My1hdcD4D3g5YmJj3LHwsnEOGKMLD2ijOjckX5nnYndauGVL3azpfYg+T9OIwRs9x5m/tZthIC+Z57B6NQUAD7zeHlqa7WhdUeiQKCeadMeY9euvdTVBRg3LochQ/oBsHLlWyxa9CfKy+caW+QpEtKYceT48K9bALj1sYls2/QpK+f/kV/NGgvAYc9h5v/2CX5xyzUAfLzhI15+9k94DniMKjciXXhGGy5ITuLODZuItVkZ1qkjN3VPpbR6O1tqD/LfF3Sl31lnsmn/AX6Vdj7T/r6ZQ4F6hnfqSFJMDIcCAaM/QkRZseItkpOTmDPnDmprDzFs2ESGDOnHRx99xh/+8Dqt+/3vJhO5Wdz4pj4YDLZmHafMhQPSGVGQA0DtnlpcZySG33vt968y4JpBJJ3ZBgCL1cpvHvpvEhITDKk1UvU58ww+9/qYdlFPZvT6MRtq9tM10cWW2oMAVNXsp9eZyfRITuRzr48b01J5ICOdA3V1CuJmuOKKAUycOCa8b7PZqK09xNy5v2PatJsNrMwAVkvjN5P53s74iy++YPbs2WzZsgW73U4wGCQtLY2pU6dy/vnnn6oaW5zNZmPpQ4vZsm4z1834FQCeWg+fvvcpv7hlWPi4tL7djSoxoiXF2Dk7Po5Z733Aj+LjmN6r53Ero47UN5Bgt5MUE0P6GclM+Ou7HG1o4IGMi9h60MOXh48YWH3kcTrjAfB6DzNhwgNMnHgthYWPM23aWGJjHQZXd4pF6zBFYWEhd9xxBxdddFH4tY0bNzJ16lTKyspavbjWlHvnGA7tH8rjtz3CpGemsHntJnoP7oPVFsF3AEzCE6hn5+Fa6kMhdh0+QiAYxGn/10O/4+02fIF6PIF6Pj3k4UDdsW74gwMHSU10KoybYffufYwffz9u91V07nwu27d/yV13leD3B9i2bQf33beQwsLToEu2RWkY19XVHRfEAL169WrNelpd1esbOFBzgCG5WThiHVisFqw2K5+++wmXjckyuryo8OGBQwxNOZc/bt9F21gHsTYbm/Yf4MIz2rCl9iB927Vl8/4DbDvkpZPLSVKMHW99Pd3bJPKXnV8ZXX7Eqamp5cYbiygquoX+/Y/9//Xll0sA2LlzD7ffPuf0CGKI3s64e/fuTJ06lYEDB5KYmIjP52PNmjV07x65f75feEk65XOX8uTtjxOsb+DqccOIccSwb+dezmx/ptHlRYUNNfv58RlJFGf2wmKB+Vu3seeIn9t6dsNusfCF7zDv7KkhCPzu08+5u8+FALy9p4YdvsPGFh+B5s9/gUOHvJSUlFFScuwv1oUL7yIuLtbgygwQuVmMJRQ68b3WUCjEqlWrqKqqwuv14nK56NOnD1lZWY1+OtLKHa+2WLHy/w1NuZKhr681uoyotjJr4Dc/fWJoHdEtrUXO0u2K5xp97Kd/vrFFrtlSvrcztlgsZGVlkZWlP99FJAK0cGe8adMm5s6dS2lpKdu3b2fKlClYLBa6devGzJkzsVqtLFu2jLKyMux2O+PGjWPw4MEcPXqUSZMm8fXXX+N0OnnwwQdp27bt915Ld6tEJGqEbNZGbyezcOFCpk+fjt/vB2D27Nnk5+ezZMkSQqEQq1evZt++fZSWllJWVsazzz5LcXExdXV1LF26lLS0NJYsWcI111xDSUnJSa+nMBaR6GFp/FZeXs7w4cPDW3l5+XGnSklJYd68eeH9Dz74gMzMTAAGDRrEO++8w+bNm+nduzcOh4PExERSUlLYunUrVVVVDBw4MHzs+vXrT1r6abcCT0SiWBNmU+Tk5JCTk3PC97Ozs9m5c2d4PxQKhe+VOZ1OPB4PXq+XxMR/LRxzOp14vd7jXv/nsSejMBaR6NGKK+us33oinM/nIykpCZfLhc/nO+71xMTE417/57EnPX/LlywiYpAmDFM0Vc+ePamsrASgoqKCjIwM0tPTqaqqwu/34/F4qK6uJi0tjT59+rBmzZrwsX379j3p+dUZi0j0aMVFH5MnT2bGjBkUFxeTmppKdnY2NpuNvLw83G43oVCIgoICYmNjyc3NZfLkyeTm5hITE8PDDz988tK/b55xS9A849alecatT/OMT4WWmWfcdeSiRh+77YVrW+SaLUWdsYhEj2hdDi0iElEiN4sVxiISPUImfE5xYymMRSR6aJhCRMQEIjeLFcYiEkUi+MshFMYiEj3UGYuImIBu4ImImIDCWETEeKHIzWKFsYhEEd3AExExAQ1TiIiYQOQ2xgpjEYkiWoEnImICGqYQETFeSJ2xiIgJ2BXGIiLGU2csImICGjMWETGByM1ihbGIRA9904eIiBkojEVETMCmMD6hoSlXtvYlTnsrswYaXcJpIs3oAuRkNJtCRMQENExxYv6Gv7X2JU5rsbZM4lNyjS4jqh3ZsRSAw/XrDK4keiXYB7TMiRTGIiLG03JoEREzaKEbeIFAgClTprBr1y6sViv33HMPdrudKVOmYLFY6NatGzNnzsRqtbJs2TLKysqw2+2MGzeOwYMHN+uaCmMRiR4tNEyxZs0a6uvrKSsrY926dTz66KMEAgHy8/Pp168fRUVFrF69ml69elFaWsry5cvx+/243W4GDBiAw+Fo8jUVxiISPVoojM8//3waGhoIBoN4vV7sdjsbN24kMzMTgEGDBrFu3TqsViu9e/fG4XDgcDhISUlh69atpKenN/maCmMRiR5NyOLy8nLKy8vD+zk5OeTk5ACQkJDArl27uPLKK6mtrWX+/Pls2LAByzdj0k6nE4/Hg9frJTExMXwOp9OJ1+ttVukKYxGJGk1ZDv3t8P2u559/nksuuYQ77riD3bt3c/311xMIBMLv+3w+kpKScLlc+Hy+417/djg3RQR/Y5SIyHdYLI3fvkdSUlI4VNu0aUN9fT09e/aksrISgIqKCjIyMkhPT6eqqgq/34/H46G6upq0tOYtDlJnLCLRo4VmU9xwww1MmzYNt9tNIBCgoKCACy+8kBkzZlBcXExqairZ2dnYbDby8vJwu92EQiEKCgqIjY1t1jUtoVAo1CLVn4AWfbQuLfpofVr00fpaatFH5yfWNPrYz2+9tEWu2VLUGYtI1IjgNR8KYxGJHgpjERETsERwGiuMRSRqWCN4fpjCWESihkVhLCJivAgepVAYi0j0iODHGSuMRSR6qDMWETEBhbGIiAlY9e3QIiLGU2csImICCmMRERNQGIuImICmtomImIA6YxERE9BsChERE1BnLCJiAgpjERETUBiLiJiAZlOIiJiA1WZ0Bc13WodxIFDP9GkL+HLXPqxWK3fNuolgMMSsu54jFAqR1j2FqYXXYbNF8BOrT7GLe3Xh3qlusnPuIb1nJ4pn3UBDQxB/XYCxBU+xt+YgcOzrcV56/k7+9HoVzyxaRVJiPM89Np4kVwKOGBuT71lE5bufGvxpzO/9zdU8VvwHnnl+Mju272Fm4bNYLBa6dOvA1OnXYrVaefD+xWx6bxsJzjgAHpl3G4mJCQZX3jo0TBGh3q7YREN9A6VLZrL+nfd5/LEXCDYEuS1/JBkZPZg+7WneevNdhlyWYXSpEeH2W4aSO/wSDh/2AzD3ruu5veh5Nn+4nZvGDOGOcUOZfM8iAO6aNIq2ya7w7064+T95a90HPPHsq3RLbc/v5t3Gz/5zmiGfI1I8/+yrvLzyHeLjYwF4+KFyxk8YTkZmD+69+/e89cZ7/Pyyvmz9cDtPLridM85INLji1hfJ34F3Wrd8nTqfQ0NDkGAwiNd7lBi7jeLHJpKR0YNAXT01NQc588wko8uMGJ9t38PoXz8S3r/u1sfZ/OF2AOw2K0f9AQCGXZVJMBjiL29tDB8775lXeGbRqmPH2m34vzlWTqzjeWcx97Fbw/sfffg5fS/uDsCAgT+hcv2HBINBduzYy713/Y4bxtzPH19ca1S5p4TF0vjNbE7rzjghIY5du/Zx9X9OprbWwxNP3Y7NZuXLXTX8+qYHcCUm0Pn89kaXGTH++OrfSOnYLrz/1d4DAPy0bzduuT6brJF30zOtIzlXDyD3lkeZlj88fOzBQ4cB+NFZbXju0fFMuvv3p7T2SHTZ5Rl8uasmvB8KhcKdoTMhDq/3CEeO+BntHsK1119OMBjk5l89RM8fdyat+3lGld2qzBiyjXVad8alv/8zAwb8hJWvzuEPL93H9KkL8PvrOLdDO/7057mMzPk5cx5cYnSZEW3E0J/y+P1jGfarh6jZ72HMiEGce05b/lw2nbwRlzJh7FVkXXoRAD/ufh6vLJ3OzIfKeLvyI4MrjzzWb301su/wURITE4iLi8Wddxnx8bE4nfFkZl7AJx9/YWCVrStqO+O8vDwCgeP/XPznv75lZWWtWtipkJTkxG4/dvs1qY2T+voGbvvvYgpn3ECnzufgdMZhNeN/tQgxetgljB0zhOxRs6g96AOg8P5//eNWWPBL9uw7yOtrNtGjWwcWPzWRvPGP8/5HO4wqOaL16JHC3/+2lYzMHqxb+z4XZ/Zg++dfMeW381n6h7sIBoO8996nDL1mgNGlthp7BLeX3xvGv/3tb5k+fTpPPvkkNlsEzxk5gbzrrqBo+kKuv/YeAoF6bssfSYcO7ZhRuICYGDtxcQ7uumes0WVGJKvVwsN3X88Xu2ooW3A7AGsrP+Le4j/82+NnTR5NXKyDuXddD8BBz2FGjX34lNUbDW6/M4dZM58n8GgDqantuezyDGw2K1f9V3+uy70Xe4yN//rFz+jStYPRpbYaqyVkdAnNZgmFQt9b/TPPPEOnTp3Iyspq1gX8DX9r1u9J48TaMolPyTW6jKh2ZMdSAA7XrzO4kuiVYG+Zbv3K195u9LGvXn7J977/9NNP88YbbxAIBMjNzSUzM5MpU6ZgsVjo1q0bM2fOxGq1smzZMsrKyrDb7YwbN47Bgwc3q/aT3sAbO1adoYhEhpYapaisrOS9995j6dKlHDlyhOeee47Zs2eTn59Pv379KCoqYvXq1fTq1YvS0lKWL1+O3+/H7XYzYMAAHA5Hk695Ws+mEJHo0lLDFG+//TZpaWmMHz8er9fLnXfeybJly8jMzARg0KBBrFu3DqvVSu/evXE4HDgcDlJSUti6dSvp6elNvqbCWESiRlOeTVFeXk55eXl4Pycnh5ycHABqa2v58ssvmT9/Pjt37mTcuHHHTx10OvF4PHi9XhIT/7WYxul04vV6m1W7wlhEooa9CWH87fD9ruTkZFJTU3E4HKSmphIbG8tXX30Vft/n85GUlITL5cLn8x33+rfDuSkieCKIiMjxLJZQo7fv07dvX9auXUsoFGLPnj0cOXKE/v37U1lZCUBFRQUZGRmkp6dTVVWF3+/H4/FQXV1NWlpas2pXZywiUaOlHqE5ePBgNmzYwIgRIwiFQhQVFdGxY0dmzJhBcXExqampZGdnY7PZyMvLw+12EwqFKCgoIDY2tlnXPOnUth9KU9tal6a2tT5NbWt9LTW1bfSbFY0+tmzwoBa5ZktRZywiUSOSF30ojEUkajTlBp7ZKIxFJGroa5dERExAwxQiIiagzlhExAQieeGEwlhEooaGKURETCBqHy4vIhJJIjiLFcYiEj00TCEiYgKaTSEiYgIaphARMQF1xiIiJmCzasxYRMRwGqYQETEBzaYQETEBjRmLiJiAwlhExARiNEwhImI8dcYiIiagMBYRMQGbwlhExHjqjEVETEDzjEVETCBGnfGJxdoyW/sSp70jO5YaXcJpIcE+wOgS5CQ0TPG9Pmn9S5zW0ij8+2qji4hq92UMAWDYqrUGVxK9XrpsYIucR8MUIiImEMmzKSL5IUciIsexWhq/NcbXX3/NpZdeSnV1Ndu3byc3Nxe3283MmTMJBoMALFu2jOHDhzNq1CjefPPN5tfe7N8UETEZu7Xx28kEAgGKioqIi4sDYPbs2eTn57NkyRJCoRCrV69m3759lJaWUlZWxrPPPktxcTF1dXXNq71ZvyUiYkK2JowZl5eXU15eHt7PyckhJycnvP/ggw8yevRoFixYAMAHH3xAZuaxCQmDBg1i3bp1WK1WevfujcPhwOFwkJKSwtatW0lPT29y7QpjEYkaTflT/7vh+20vvvgibdu2ZeDAgeEwDoVCWCzHxjecTicejwev10tiYmL495xOJ16vt1m1K4xFJGq01NS25cuXY7FYWL9+PR999BGTJ09m//794fd9Ph9JSUm4XC58Pt9xr387nJtCY8YiEjVa6gbe4sWLWbRoEaWlpVxwwQU8+OCDDBo0iMrKSgAqKirIyMggPT2dqqoq/H4/Ho+H6upq0tLSmlW7OmMRiRpNGTNuqsmTJzNjxgyKi4tJTU0lOzsbm81GXl4ebrebUChEQUEBsbGxzTq/wlhEokZjZkk0VWlpafjnRYsW/b/3R40axahRo37wdRTGIhI1tBxaRMQEInkFnsJYRKKGnk0hImICkTw9TGEsIlFDY8YiIiYQY9UwhYiI4dQZi4iYgMJYRMQEdANPRMQELOqMRUSMp2EKERET0DCFiIgJWLQCT0TEeBE8SqEwFpHooRt4IiImEMFZrDAWkeihR2iKiJiAhilEREwggrNYYSwi0UNhLCJiAlqBJyJiAhGcxad3GAcC9Uyb9hi7du2lri7AuHE5dO16HlOmPIrFYqFbt07MnHkLVmskL7I0ztGDHlZPn83AqRMI1gV497mlWGw2EtufTd+xY7BYrXz88ut88c7fsVgt9PjFFXS4uJfRZUeMz2bfjTU+HoCYM9vR9tIh7C4rxWKPIa7jeZwzYjQWq5Wvli3h8GfbsMbFAXDeb27FFp9gZOmtRt+BF6FWrHiL5OQk5sy5g9raQwwbNpEePVLJz8+jX7+fUFT0JKtXV5KV1d/oUiNOsL6Bd59bgs3hAODDl17hguFX0b7XhVQ++T/s3riFdt27su0vb3Fl8d3U++tYNfV+hXEjBQMBADrn3xl+7bMH7+GckbkkpHZl78qXOPj3SpIz+3Pkix2k3FqA3ZVoVLmnTCTPpmhyy1dXV9cadRjiiisGMHHimPC+zWbjgw+2kZl5IQCDBvXlnXc2GlRdZNu8ZDmpQwYSd0YbAJI7nUed10coFKL+6FGsNhv22FgS2rWl3l9H/VF/ZA/4nWJHd35BsK6O7fOK+fyxuRz+RzWB2loSUrsCEJ/alSPV2wgFg9Tt28PuJb/nHw/Ppvadtw2uvHVZm7CZzQlreuONNxg8eDBZWVm88sor4dfHjh17Sgo7FZzOeFyuBLzew0yY8AD5+dcSCoHlm39enc54PJ7DBlcZeT5fs57YxETOSe8Zfs11zlls/P0LvDZpFkcPejjrgjQAEtqewWt3zmL19Nl0zf4PgyqOPFaHgzMvyybl1gLaj76WXc8/g6PdWfg+/RgA7/ubCPr9BOvqaHvpEDrcMJaU8QXUrn2To7u+MLj61mOxNH4zmxMOU8yfP5+XXnqJUCjExIkT8fv9DBs2jFAocsdk/p3du/cxfvz9uN1XMXTofzBnzvPh93y+IyQlOY0rLkJ9vmY9WGDPB1s5uH0nG576HQe27+Sy+6fSpuO5bHttDZsXL+dH6T05euAgVz5yDwBrH5xHu7QutO3S2dgPEAEcZ/8Ix1lnY7FYiP3ROdicTs6+ejg1f3mFr1//M3GdOmOx27E6HLQdfBlWRywAzrQeHN25k7gO5xn8CVqHCTO20U4YxjExMSQnJwNQUlLC9ddfT/v27cNdYzSoqanlxhuLKCq6hf79LwKgZ89UKivfp1+/n1BRUcVPf5pucJWR5z+Kbg///Na9j9DnxlzWFz9NTPyxG0jxZ7Th60+qiXEmYHM4sMbYsVgsxCQkUOfTXyKNcWD92/i/3EX70dcSOHCA4NGjHPn8M8699lfEJCeze9kSXD0vpG7vV+x8bgGpU4ogFORw9Tba9PuZ0eW3mpYa6QoEAkybNo1du3ZRV1fHuHHj6Nq1K1OmTPnm5n43Zs6cidVqZdmyZZSVlWG32xk3bhyDBw9u1jVPGMYdOnRg9uzZTJw4EZfLxRNPPMFNN93EoUOHmv0BzWb+/Bc4dMhLSUkZJSVlABQW/pp7732a4uJ6UlPPIzs7ev+Heyr1vXkMlfOew2KzYrXb6Tt2DM6zzmTvlq28MXMOFouFdt278KOfXGB0qRHhjJ8NZFfpc/zj4QfAYuHca2+gwedlR8ljWB0OnGndSbzwWCPR5uJ+/GPO/VhsNtr060/cuR0Mrr71tFQYr1ixguTkZObMmUNtbS3Dhg2jR48e5Ofn069fP4qKili9ejW9evWitLSU5cuX4/f7cbvdDBgwAMc3N66bwhI6wbhDfX09K1as4MorryT+m+kzNTU1PP300xQWFjbhEp80uShpijQK/77a6CKi2n0ZQwAYtmqtwZVEr5cuG9gi59l9eGWjj22fMPSE7/l8x242u1wuamtrGTFiBHV1dVRUVGCxWFi1ahXr1q3jkksuYc2aNcyaNQuA8ePH85vf/Ib09Kb/RX3CG3h2u53hw4eHgxigXbt2TQxiEZFTx2IJNXorLy9n+PDh4a28vDx8HqfTicvlwuv1MmHCBPLz8wmFQt+6ue/E4/Hg9XpJTEw87ve8Xm+zaj+t5xmLSHRpyihFTk4OOTk5J3x/9+7djB8/HrfbzdChQ5kzZ074PZ/PR1JSEi6XC5/Pd9zr3w7npjDjdDsRkWZpqaltNTU13HjjjUyaNIkRI0YA0LNnTyorKwGoqKggIyOD9PR0qqqq8Pv9eDweqqurSUtLa1bt6oxFJGrYWug88+fP59ChQ5SUlFBSUgJAYWEh9957L8XFxaSmppKdnY3NZiMvLw+3200oFKKgoIDY2NhmXfOEN/Bajm7gtS7dwGttuoHX+lrqBt5+/4pGH9s29hctcs2Wos5YRKJI5K6DUBiLSNSwKIxFRIxnsUTunASFsYhEEXXGIiKGs0TwbF2FsYhEDQ1TiIiYgoYpREQMp9kUIiImoDAWETEBi6WlFkSfegpjEYki6oxFRAynYQoREVPQ1DYREcOpMxYRMYFI/vZ6hbGIRA1Liz1e/tRTGItIFFFnLCJiOA1TiIiYgsJYRMRweoSmiIgpqDMWETGcVc8zFhExA4WxiIjhtAJPRMQUFMYiIoaL5HnGllAoFDK6CBGRlvFJE45Na7UqmkNhLCJiApF761FEJIoojEVETEBhLCJiAgpjERETUBiLiJiAwlhExAQUxiIiJqAwBoLBIEVFReTk5JCXl8f27duNLilqbdq0iby8PKPLiFqBQIBJkybhdrsZMWIEq1evNrokaSQthwZWrVpFXV0d5eXlbNy4kQceeICnnnrK6LKizsKFC1mxYgXx8fFGlxK1VqxYQXJyMnPmzKG2tpZhw4YxZMgQo8uSRlBnDFRVVTFw4EAAevXqxZYtWwyuKDqlpKQwb948o8uIaldccQUTJ04M79tskfttyacbhTHg9XpxuVzhfZvNRn19vYEVRafs7Gzsdv0x1pqcTiculwuv18uECRPIz883uiRpJIUx4HK58Pl84f1gMKjQkIi1e/durrvuOq6++mqGDh1qdDnSSApjoE+fPlRUVACwceNG0tLM9TQnkcaqqanhxhtvZNKkSYwYMcLocqQJ1P4BWVlZrFu3jtGjRxMKhbj//vuNLkmkWebPn8+hQ4coKSmhpKQEOHbjNC4uzuDK5GT0CE0RERPQMIWIiAkojEVETEBhLCJiAgpjERETUBiLiJiAwlhaxIIFC7jkkkvw+/0nPObjjz9mw4YNTT73lClTwvPARaKVwlhaxMqVK7nqqqt4+eWXT3jMa6+9xrZt205hVSKRQ2EsP1hlZSUpKSmMHj2axYsXA8celTlq1ChGjhzJrbfeyp49e3jppZd4/vnn2bx5Mz//+c/DXfTcuXN58cUXaWhooLCwkJtuuonhw4fz6KOPGvipRE4trcCTH+yFF15g5MiRpKam4nA42LRpEzNmzOCRRx6hS5cuLF68mJqaGoYNG0a7du1IT0//t+fZvXs3vXr1YuTIkfj9fgYNGqQH3chpQ2EsP8jBgwepqKhg//79lJaW4vV6WbRoEV9//TVdunQBYMyYMQC88cYb//Yc/1wEmpyczPvvv89f//pXXC4XdXV1p+ZDiJiAwlh+kBUrVvDLX/6SyZMnA3DkyBGGDBlCXFwcn3/+OZ07d2bBggWcf/75WCwWgsEgAA6Hg71799KxY0e2bt1Kly5dePHFF0lMTGTWrFls376dZcuWodX6crpQGMsP8sILL/DQQw+F9+Pj47n88stp164d06ZNw2q1ctZZZ3HDDTcQExPDQw89RJcuXRg7diy//vWv6dChA0lJSQD079+f22+/naqqKuLj4+nUqRN79+416qOJnFJ6UJCIiAloNoWIiAkojEVETEBhLCJiAgpjERETUBiLiJiAwlhExAQUxiIiJvB/zaljKSR5+MwAAAAASUVORK5CYII=",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Create a pipeline which transforms phrases into normalized feature vectors and uses a bayes estimator\n",
    "model_name = 'bayes classifier'\n",
    "\n",
    "pipeline_bayes = Pipeline([\n",
    "                ('count', CountVectorizer()),\n",
    "                ('tfidf', TfidfTransformer()),\n",
    "                ('gnb', MultinomialNB()),\n",
    "                ])\n",
    "\n",
    "# Train model using the created sklearn pipeline\n",
    "model_bayes = pipeline_bayes.fit(train_df['text'], train_df['sentiment'])\n",
    "\n",
    "# Evaluate model performance\n",
    "model_score = evaluate_results(model_bayes, test_df)\n",
    "performance_df = performance_df.append({'model_name': model_name, \n",
    "                                    'f1_score': model_score[0], \n",
    "                                    'precision': model_score[1], \n",
    "                                    'recall': model_score[2]}, ignore_index=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step #5 Compare Classifier Performance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                       model_name  f1_score  precision    recall\n",
      "1                bayes classifier  0.592866   0.712067  0.604659\n",
      "0  logistic regression classifier  0.701729   0.725093  0.709935\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1a8a608cca0>"
      ]
     },
     "execution_count": 99,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Compare model performance\n",
    "print(performance_df)\n",
    "\n",
    "performance_df = performance_df.sort_values('model_name')\n",
    "fig, ax = plt.subplots(figsize=(6, 4))\n",
    "tidy = performance_df.melt(id_vars='model_name').rename(columns=str.title)\n",
    "sns.barplot(x='Model_Name', y='Value', hue='Variable', data=tidy, ax=ax, linewidth=1, edgecolor=\"w\")\n",
    "plt.title('Model Outlier Detection Performance (Macro)')\n",
    "plt.legend(bbox_to_anchor=(1.02, 1), loc='upper left', borderaxespad=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Step #6 Make Test Predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mondays just suck!-> Negative\n",
      "I love this product-> Positive\n",
      "That is a tree-> Neutral\n",
      "Terrible service-> Negative\n"
     ]
    }
   ],
   "source": [
    "testphrases = ['Mondays just suck!', 'I love this product', 'That is a tree', 'Terrible service']\n",
    "for testphrase in testphrases:\n",
    "    resultx = model_lgr.predict([testphrase]) # use model_bayes for predictions with the other model\n",
    "    dict = {1: 'Negative', 2: 'Neutral', 3: 'Positive'}\n",
    "    print(testphrase + '-> ' + dict[resultx[0]])"
   ]
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "38b6c26bd9870504e27f7883362b971e6afb0a02e23f00032a9076a16838eb05"
  },
  "kernelspec": {
   "display_name": "Python 3.8.8 ('base')",
   "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}