sentiment-analysis-model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "994d6af3-4290-4ae5-b7a4-10c25a9f9943",
   "metadata": {
    "id": "994d6af3-4290-4ae5-b7a4-10c25a9f9943"
   },
   "source": [
    "# Task 1 : Create a DataFrame from a Text File with Sentence-Label Pairs"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38b1c426-3cfd-42a6-9da9-321a9d66b2c3",
   "metadata": {
    "id": "38b1c426-3cfd-42a6-9da9-321a9d66b2c3"
   },
   "source": [
    "<div class=\"task-panel-v2--description--15n8T\">\n",
    "    <div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\"\n",
    "        class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\">\n",
    "        <p>Construct a Pandas DataFrame from a text file that contains sentence-label pairs. Each line in the text file\n",
    "            follows the format <code><strong>sentence@label</strong></code>. This task involves:</p>\n",
    "        <ul>\n",
    "            <li>\n",
    "                <p>Reading data from 'Sentences_75Agree_sample.txt'.</p>\n",
    "            </li>\n",
    "            <li>\n",
    "                <p>Parsing each line to separate sentences from labels.</p>\n",
    "            </li>\n",
    "            <li>\n",
    "                <p>Creating a DataFrame with two columns: 'Sentence' and 'Label'.</p>\n",
    "                <p><br></p>\n",
    "            </li>\n",
    "        </ul>\n",
    "    </div>\n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "e819d824-252a-4df0-9612-c24190504a8e",
   "metadata": {
    "id": "e819d824-252a-4df0-9612-c24190504a8e"
   },
   "outputs": [],
   "source": [
    "# import pandas for data manipulation\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "fd02a5d9-0ddc-462d-adf3-096d657811f3",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "fd02a5d9-0ddc-462d-adf3-096d657811f3",
    "outputId": "0899ad2b-5b6c-44a7-c1d9-96c319953e1a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['According to Gran , the company has no plans to move all production to Russia , although that is where the company is growing .@neutral\\n',\n",
       " 'With the new production plant the company would increase its capacity to meet the expected increase in demand and would improve the use of raw materials and therefore increase the production profitability .@positive\\n',\n",
       " \"For the last quarter of 2010 , Componenta 's net sales doubled to EUR131m from EUR76m for the same period a year earlier , while it moved to a zero pre-tax profit from a pre-tax loss of EUR7m .@positive\\n\",\n",
       " 'In the third quarter of 2010 , net sales increased by 5.2 % to EUR 205.5 mn , and operating profit by 34.9 % to EUR 23.5 mn .@positive\\n',\n",
       " 'Operating profit rose to EUR 13.1 mn from EUR 8.7 mn in the corresponding period in 2007 representing 7.7 % of net sales .@positive\\n']"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Open the text file and read all lines\n",
    "with open('Sentences_75Agree_sample.txt', 'r', encoding='latin1') as file:\n",
    "    lines = file.readlines()\n",
    "lines[:5]\n",
    "\n",
    "# To close the file automatically without calling the close() method, you use the 'with' statement as here\n",
    "# '''\n",
    "# https://www.pythontutorial.net/python-basics/python-read-text-file/\n",
    "# The code in the previous examples works fine with ASCII text files. However, if you’re dealing with other languages such as Japanese, Chinese, and\n",
    "# Korean, the text file is not a simple ASCII text file. And it’s likely a UTF-8 file that uses more than just the standard ASCII text characters.\n",
    "\n",
    "# To open a UTF-8 text file, you need to pass the encoding='utf-8' to the open() function to instruct it to expect UTF-8 characters from the file.\n",
    "# '''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "39c45138-3382-463c-81ad-236db565e977",
   "metadata": {
    "id": "39c45138-3382-463c-81ad-236db565e977"
   },
   "outputs": [],
   "source": [
    "# Split each linr at '@' to separate sentences and labels and strip whitespaces\n",
    "data = [line.strip().split('@') for line in lines]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2efe112d-c4de-44c8-9928-5f5eb9de1656",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "2efe112d-c4de-44c8-9928-5f5eb9de1656",
    "outputId": "19d50eb6-3641-4317-e334-04d564592ae0"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[['According to Gran , the company has no plans to move all production to Russia , although that is where the company is growing .',\n",
       "  'neutral'],\n",
       " ['With the new production plant the company would increase its capacity to meet the expected increase in demand and would improve the use of raw materials and therefore increase the production profitability .',\n",
       "  'positive'],\n",
       " [\"For the last quarter of 2010 , Componenta 's net sales doubled to EUR131m from EUR76m for the same period a year earlier , while it moved to a zero pre-tax profit from a pre-tax loss of EUR7m .\",\n",
       "  'positive']]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[:3]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "1432909e-a81d-421a-b794-9b3c584ed46b",
   "metadata": {
    "id": "1432909e-a81d-421a-b794-9b3c584ed46b"
   },
   "outputs": [],
   "source": [
    "df = pd.DataFrame(data, columns=['Sentence', 'Label'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "db040921-3674-4d04-9ef8-57563fa48864",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "db040921-3674-4d04-9ef8-57563fa48864",
    "outputId": "0dce79cd-5980-454c-948c-b05c61fd8420"
   },
   "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>Sentence</th>\n",
       "      <th>Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>According to Gran , the company has no plans t...</td>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>With the new production plant the company woul...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>For the last quarter of 2010 , Componenta 's n...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>In the third quarter of 2010 , net sales incre...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Operating profit rose to EUR 13.1 mn from EUR ...</td>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            Sentence     Label\n",
       "0  According to Gran , the company has no plans t...   neutral\n",
       "1  With the new production plant the company woul...  positive\n",
       "2  For the last quarter of 2010 , Componenta 's n...  positive\n",
       "3  In the third quarter of 2010 , net sales incre...  positive\n",
       "4  Operating profit rose to EUR 13.1 mn from EUR ...  positive"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "29548b3a-72f8-4e86-befb-107c7bc8d5ae",
   "metadata": {
    "id": "29548b3a-72f8-4e86-befb-107c7bc8d5ae"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "71c47d1c-5873-46a8-baf4-37d2e419300b",
   "metadata": {
    "id": "71c47d1c-5873-46a8-baf4-37d2e419300b"
   },
   "source": [
    "# Task 2: Visual Analysis of Label Distribution and Sentence Lengths in Text Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32ec75df-5330-404d-b76f-868047a42051",
   "metadata": {
    "id": "32ec75df-5330-404d-b76f-868047a42051"
   },
   "source": [
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>In this task, you will analyze a preloaded dataset containing sentences and their corresponding labels. You will visualize the frequency of each label to understand the distribution of categories within the data. Additionally, you will calculate and plot the distribution of sentence lengths to gain insights into the variability of sentence construction within the dataset.</p><p><br></p></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ade6f3cd-9bf7-4e44-b02b-c42ea375f6d6",
   "metadata": {
    "id": "ade6f3cd-9bf7-4e44-b02b-c42ea375f6d6"
   },
   "outputs": [],
   "source": [
    "# Pandas already imported and Matplotlib to be imported\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "d3e3ff9a-51eb-44f7-ad73-5bbebe03d904",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "d3e3ff9a-51eb-44f7-ad73-5bbebe03d904",
    "outputId": "831c8fd7-489e-42af-e18c-4024a54e2055"
   },
   "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>Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>neutral</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>positive</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Label\n",
       "0   neutral\n",
       "1  positive\n",
       "2  positive\n",
       "3  positive\n",
       "4  positive"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create Df for 'Label' column for Analysis and easy mannipulation\n",
    "labels_df = pd.DataFrame(df.Label, columns=['Label'])\n",
    "labels_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "42814d2a-5d3b-4a4f-a92d-800be78ea71d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 519
    },
    "id": "42814d2a-5d3b-4a4f-a92d-800be78ea71d",
    "outputId": "a3d26d54-e7a8-4990-9bb3-b663ce1721dd"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use a bar chart to visualize how frequently each labe occurs in the dataset\n",
    "labels_df['Label'].value_counts().plot(kind='bar')\n",
    "plt.title('Distribution of Labels')\n",
    "plt.xlabel('Label')\n",
    "plt.ylabel('Count')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2a76a95b-2b39-40a7-92e1-e997387ac72b",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "2a76a95b-2b39-40a7-92e1-e997387ac72b",
    "outputId": "bd28ee30-ed24-49f0-dc72-352d378db629"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[25, 33, 39, 29, 24]"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Determine the number of words in each sentence to analyze sentence length\n",
    "sentence_lengths = [len(sentence.split()) for sentence in df.Sentence]\n",
    "sentence_lengths[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "0dad84ca-6eef-4131-bf44-9a317af17e60",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 489
    },
    "id": "0dad84ca-6eef-4131-bf44-9a317af17e60",
    "outputId": "4c37a4e6-c2b3-4fb2-e1ce-867572b8244a"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot a histogram to show the distribution of sentence lengths in dataset\n",
    "plt.hist(sentence_lengths, bins=30)\n",
    "plt.title('Distribution of Sentence Lengths')\n",
    "plt.xlabel('Sentence Length')\n",
    "plt.ylabel('Frequency')\n",
    "plt.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f646e4f1-3886-48ed-91a7-3fd677e2c73f",
   "metadata": {
    "id": "f646e4f1-3886-48ed-91a7-3fd677e2c73f"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "1014641c-1c74-4561-8fb5-44f88f4fc7a3",
   "metadata": {
    "id": "1014641c-1c74-4561-8fb5-44f88f4fc7a3"
   },
   "source": [
    "# Task 3: Visualize Top 20 Most Frequent Words in Text Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f7e12919-5b1f-43fc-839d-861c5b8d02d9",
   "metadata": {
    "id": "f7e12919-5b1f-43fc-839d-861c5b8d02d9"
   },
   "source": [
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>This task involves processing text data to extract and visualize the most frequently occurring words. By tokenizing the sentences and counting the occurrences of each word, you will identify the top 20 most common words in the dataset. Then, you will create a bar plot to visualize these words and their frequencies to understand the common vocabulary within your text data.</p><p><br></p></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "52bd9bcc-b987-47b8-938d-3b85fa82d6b2",
   "metadata": {
    "id": "52bd9bcc-b987-47b8-938d-3b85fa82d6b2"
   },
   "outputs": [],
   "source": [
    "# Import Counter for frequency analysis and seaborn for advanced plotting\n",
    "from collections import Counter\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "a5268a29-a761-43d6-b3a8-55492e38eacf",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "a5268a29-a761-43d6-b3a8-55492e38eacf",
    "outputId": "4d72d204-adfb-463b-c27b-f0b10957d6f8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['According', 'to', 'Gran', ',', 'the', 'company', 'has', 'no', 'plans', 'to']"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Flatten the list of sentences into words\n",
    "words = [word for sentence in df.Sentence for word in sentence.split()]\n",
    "words[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "0cb8d9b9-69f0-4843-b8e1-451b4eb33cef",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "0cb8d9b9-69f0-4843-b8e1-451b4eb33cef",
    "outputId": "a5879f56-7ef4-4076-b6b6-11b21ece8c94"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('.', 1463), (',', 1461), ('the', 1404), ('of', 1016), ('to', 901)]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Utilize Counter to find top 20 most common words\n",
    "common_words = Counter(words).most_common(20)\n",
    "common_words[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "b3db991d-ac6a-4f3e-884e-f6af41fbac62",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "b3db991d-ac6a-4f3e-884e-f6af41fbac62",
    "outputId": "0f136752-34b7-4831-a3a3-3373b244b868"
   },
   "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>Word</th>\n",
       "      <th>Frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>.</td>\n",
       "      <td>1463</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>,</td>\n",
       "      <td>1461</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>the</td>\n",
       "      <td>1404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>of</td>\n",
       "      <td>1016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>to</td>\n",
       "      <td>901</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  Word  Frequency\n",
       "0    .       1463\n",
       "1    ,       1461\n",
       "2  the       1404\n",
       "3   of       1016\n",
       "4   to        901"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a Dataframe from the most common words for easy plotting with seaborn\n",
    "common_words_df = pd.DataFrame(common_words, columns=['Word', 'Frequency'])\n",
    "common_words_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "3daea589-262c-4aed-86d1-25e89975c3ea",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 489
    },
    "id": "3daea589-262c-4aed-86d1-25e89975c3ea",
    "outputId": "dfd0432e-953a-4e74-9577-1bc205316aa3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use Seaborn's barplot to create horizontal bar plot\n",
    "sns.barplot(data=common_words_df, x='Frequency', y='Word')\n",
    "plt.title('Most Coomon Words')\n",
    "plt.plot()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bb34be5b-ce9f-4cd2-bae1-a0b50ae29455",
   "metadata": {
    "id": "bb34be5b-ce9f-4cd2-bae1-a0b50ae29455"
   },
   "source": [
    "# Task 4: Analyze and Visualize Most Frequent Bi-grams in Text Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8e6be0a5-0894-4834-9945-618de991f6a8",
   "metadata": {
    "id": "8e6be0a5-0894-4834-9945-618de991f6a8"
   },
   "source": [
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>In this task, you will extract bi-grams from a corpus of sentences to discover common phrases or pairs of words. Using the CountVectorizer from the scikit-learn library, you will calculate the frequency of each bi-gram in the dataset. Then, you will sort these bi-grams by their frequency and visualize the top 20 most frequent bi-grams.</p><p><br></p></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2910cf40-979c-49dd-bf1b-32b6a91e88f1",
   "metadata": {
    "id": "2910cf40-979c-49dd-bf1b-32b6a91e88f1"
   },
   "outputs": [],
   "source": [
    "# '''\n",
    "# In the Bigram Language Model, we find bigrams, which are two words coming together in the corpus(the entire collection of words/sentences).\n",
    "\n",
    "# For example: In the sentence, Edpresso is awesome, and user-friendly the bigrams are :\n",
    "\n",
    "# “Edpresso is”\n",
    "# “is awesome”\n",
    "# “and user”\n",
    "# “user friendly”\n",
    "# '''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b2a6f5a8-2f65-48f0-bb04-eba7eaaf767c",
   "metadata": {
    "id": "b2a6f5a8-2f65-48f0-bb04-eba7eaaf767c"
   },
   "outputs": [],
   "source": [
    "# Import CountVectorizer for text feature extraction and Seaborn for plotting\n",
    "from sklearn.feature_extraction.text import CountVectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "872a805f-1074-4e24-b075-2ca566398061",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 75
    },
    "id": "872a805f-1074-4e24-b075-2ca566398061",
    "outputId": "6966b214-23e1-49e5-bd2e-c1a5fa47c130"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>#sk-container-id-1 {color: black;background-color: white;}#sk-container-id-1 pre{padding: 0;}#sk-container-id-1 div.sk-toggleable {background-color: white;}#sk-container-id-1 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-1 label.sk-toggleable__label-arrow:before {content: \"▸\";float: left;margin-right: 0.25em;color: #696969;}#sk-container-id-1 label.sk-toggleable__label-arrow:hover:before {color: black;}#sk-container-id-1 div.sk-estimator:hover label.sk-toggleable__label-arrow:before {color: black;}#sk-container-id-1 div.sk-toggleable__content {max-height: 0;max-width: 0;overflow: hidden;text-align: left;background-color: #f0f8ff;}#sk-container-id-1 div.sk-toggleable__content pre {margin: 0.2em;color: black;border-radius: 0.25em;background-color: #f0f8ff;}#sk-container-id-1 input.sk-toggleable__control:checked~div.sk-toggleable__content {max-height: 200px;max-width: 100%;overflow: auto;}#sk-container-id-1 input.sk-toggleable__control:checked~label.sk-toggleable__label-arrow:before {content: \"▾\";}#sk-container-id-1 div.sk-estimator input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 div.sk-label input.sk-toggleable__control:checked~label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 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-1 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-1 div.sk-estimator:hover {background-color: #d4ebff;}#sk-container-id-1 div.sk-parallel-item::after {content: \"\";width: 100%;border-bottom: 1px solid gray;flex-grow: 1;}#sk-container-id-1 div.sk-label:hover label.sk-toggleable__label {background-color: #d4ebff;}#sk-container-id-1 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-1 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-1 div.sk-item {position: relative;z-index: 1;}#sk-container-id-1 div.sk-parallel {display: flex;align-items: stretch;justify-content: center;background-color: white;position: relative;}#sk-container-id-1 div.sk-item::before, #sk-container-id-1 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-1 div.sk-parallel-item {display: flex;flex-direction: column;z-index: 1;position: relative;background-color: white;}#sk-container-id-1 div.sk-parallel-item:first-child::after {align-self: flex-end;width: 50%;}#sk-container-id-1 div.sk-parallel-item:last-child::after {align-self: flex-start;width: 50%;}#sk-container-id-1 div.sk-parallel-item:only-child::after {width: 0;}#sk-container-id-1 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-1 div.sk-label label {font-family: monospace;font-weight: bold;display: inline-block;line-height: 1.2em;}#sk-container-id-1 div.sk-label-container {text-align: center;}#sk-container-id-1 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-1 div.sk-text-repr-fallback {display: none;}</style><div id=\"sk-container-id-1\" class=\"sk-top-container\"><div class=\"sk-text-repr-fallback\"><pre>CountVectorizer(ngram_range=(2, 2))</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-1\" type=\"checkbox\" checked><label for=\"sk-estimator-id-1\" class=\"sk-toggleable__label sk-toggleable__label-arrow\">CountVectorizer</label><div class=\"sk-toggleable__content\"><pre>CountVectorizer(ngram_range=(2, 2))</pre></div></div></div></div></div>"
      ],
      "text/plain": [
       "CountVectorizer(ngram_range=(2, 2))"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Set up CountVectorizer to consider bi0grams only\n",
    "vectorizer = CountVectorizer(ngram_range=(2,2))\n",
    "vectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "a9d75c2b-d9ee-4401-9483-13063b6ef50c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "a9d75c2b-d9ee-4401-9483-13063b6ef50c",
    "outputId": "22ca95f1-34ef-47ab-c993-7870b761c7f1"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<1492x17758 sparse matrix of type '<class 'numpy.int64'>'\n",
       "\twith 28938 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Apply the vectorizer to the sentences to get the bi-gram counts\n",
    "ngram_counts = vectorizer.fit_transform(df.Sentence)\n",
    "ngram_counts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "8c4ce73b-febc-449d-b25a-d49e5b194c65",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8c4ce73b-febc-449d-b25a-d49e5b194c65",
    "outputId": "6e40b0fe-e14a-4d3b-f9b7-7636388eaaaf"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('according to', 30),\n",
       " ('to gran', 1),\n",
       " ('gran the', 1),\n",
       " ('the company', 165),\n",
       " ('company has', 12)]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Aggregate counts and get feature names (bi-grams)\n",
    "ngram_sum = ngram_counts.sum(axis=0)\n",
    "ngram_list = [(ngram, ngram_sum[0, idx]) for ngram, idx in vectorizer.vocabulary_.items()]\n",
    "ngram_list[:5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "18ccc95a-9af6-4c4a-a3fb-5e00ba4e9139",
   "metadata": {
    "id": "18ccc95a-9af6-4c4a-a3fb-5e00ba4e9139"
   },
   "outputs": [],
   "source": [
    "# Sort the bi-grams by their frequency count in descending order\n",
    "common_ngrams = sorted(ngram_list, key=lambda x: x[1], reverse=True)[:20]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "1c9a552a-7b31-4ca5-ae94-2346845de77f",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "1c9a552a-7b31-4ca5-ae94-2346845de77f",
    "outputId": "25b28b35-7216-445a-ccbb-a0885b6a7bea"
   },
   "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>N-gram</th>\n",
       "      <th>Frequency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>in the</td>\n",
       "      <td>264</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>of the</td>\n",
       "      <td>186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>for the</td>\n",
       "      <td>175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>the company</td>\n",
       "      <td>165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>net sales</td>\n",
       "      <td>143</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        N-gram  Frequency\n",
       "0       in the        264\n",
       "1       of the        186\n",
       "2      for the        175\n",
       "3  the company        165\n",
       "4    net sales        143"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a DataFrame for the most common bi-grams to facilitate visualization\n",
    "common_ngrams_df = pd.DataFrame(common_ngrams, columns=['N-gram', 'Frequency'])\n",
    "common_ngrams_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "dd8cac85-3d3b-4c74-88b4-0b36c3d16e89",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 472
    },
    "id": "dd8cac85-3d3b-4c74-88b4-0b36c3d16e89",
    "outputId": "b2fca850-16c0-48d4-84b8-ff9102799add"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use seaborn to create a bar plot for the bi-gram frequencies\n",
    "sns.barplot(data=common_ngrams_df, x='Frequency', y='N-gram')\n",
    "plt.title('Most Common Bi-grams')  # Adding a title to the plot\n",
    "plt.show()                          # Display the visualization"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "830b8ddf-6e75-44f9-ab12-b94d938d892e",
   "metadata": {
    "id": "830b8ddf-6e75-44f9-ab12-b94d938d892e"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "bfa60779-7e8a-4f9a-8d9b-8a2b2621d7a0",
   "metadata": {
    "id": "bfa60779-7e8a-4f9a-8d9b-8a2b2621d7a0"
   },
   "source": [
    "# Task 5: Analyze Sentiment Distribution Using Positive and Negative Word Counts"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ca034870-ed81-487f-915b-4fbd86703e85",
   "metadata": {
    "id": "ca034870-ed81-487f-915b-4fbd86703e85"
   },
   "source": [
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>In this task, you will conduct a simple sentiment analysis by tallying the occurrences of predefined positive and negative sentiment words in a collection of sentences. The goal is to obtain a rough estimate of the sentiment distribution within the text data. To complete the task:</p><ul><li><p>Create two lists: one containing words that are typically associated with 6 positive sentiments (e.g., good, great, positive), and another with 6 words associated with negative sentiments&nbsp; (e.g., bad, poor, down).</p></li><li><p>Calculate the frequency of Sentiment Words</p></li><li><p>Plot the Counts of Sentiment Words</p><p><br></p></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "0bdfa217-c1fc-4d52-84ba-2b4f5b9530a0",
   "metadata": {
    "id": "0bdfa217-c1fc-4d52-84ba-2b4f5b9530a0"
   },
   "outputs": [],
   "source": [
    "# Create lists of words associated with positive and negative sentiments\n",
    "positive_words = ['good', 'great', 'positive', 'profit', 'up', 'increase']\n",
    "negative_words = ['bad', 'poor', 'negative', 'loss', 'down', 'decrease']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "0dd86b9c-69cc-4174-b963-9a567802f17d",
   "metadata": {
    "id": "0dd86b9c-69cc-4174-b963-9a567802f17d"
   },
   "outputs": [],
   "source": [
    "# Calculate the frequency of positive and negative words in the dataset\n",
    "positive_counts = sum([sentence.lower().count(word) for sentence in df['Sentence'] for word in positive_words])\n",
    "negative_counts = sum([sentence.lower().count(word) for sentence in df['Sentence'] for word in negative_words])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "e8846137-e049-4c3c-b079-1245becd53d6",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 452
    },
    "id": "e8846137-e049-4c3c-b079-1245becd53d6",
    "outputId": "4da2876c-9480-4433-8188-3d323c7469df"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Use seaborn to create a bar plot showing the counts of positive and negative sentiment words\n",
    "sns.barplot(x=['Positive Words', 'Negative Words'], y=[positive_counts, negative_counts])\n",
    "plt.title('Occurrence of Sentiment Words')  # Set the title of the plot\n",
    "plt.ylabel('Count')                          # Set the y-axis label\n",
    "plt.show()                                   # Display the plot"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dca13444-22f8-45ed-b08a-21964419e938",
   "metadata": {
    "id": "dca13444-22f8-45ed-b08a-21964419e938"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "1676e9ec-9122-4b58-b12d-a931b4af6a3d",
   "metadata": {
    "id": "1676e9ec-9122-4b58-b12d-a931b4af6a3d"
   },
   "source": [
    "# Task 6: Text Data Preprocessing for Sentiment Analysis Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "61149cb8-7198-47dd-b721-b78aa7d70238",
   "metadata": {
    "id": "61149cb8-7198-47dd-b721-b78aa7d70238"
   },
   "source": [
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>Your task is to clean and transform sentences from a dataset so that they can be used by a machine learning model. Specifically, you need to:</p><ul><li><p>Clean the sentences in the dataset.</p></li><li><p>Convert the cleaned sentences into a format suitable for machine learning (TF-IDF).</p></li><li><p>Change text labels into numbers.</p></li><li><p>Split the data into training and testing sets, with a&nbsp; test size of 20% and a random state set to 1502.</p></li><li><p>Make sure the data is in the right format for the machine learning model to process.</p><p><br></p></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "7b06f65b-e499-4f52-a450-3db5468d2055",
   "metadata": {
    "id": "7b06f65b-e499-4f52-a450-3db5468d2055"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "851be2c9-f40f-48eb-b1f5-fc6466ec3792",
   "metadata": {
    "id": "851be2c9-f40f-48eb-b1f5-fc6466ec3792"
   },
   "outputs": [],
   "source": [
    "# Start by defining a function to clean the text data, aiming to standardize it for analysis.\n",
    "import re\n",
    "\n",
    "# Function to clean text data\n",
    "def clean_text(text):\n",
    "    text = re.sub(r\"[^\\w\\s]\", '', text)  # Remove punctuation\n",
    "    text = text.lower()  # Convert to lowercase\n",
    "    text = re.sub(r\"\\s+\", ' ', text)  # Collapse multiple spaces to one\n",
    "    return text.strip()  # Remove leading and trailing spaces"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "39e6fae5-6778-494e-837c-a42761c7b9b3",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "39e6fae5-6778-494e-837c-a42761c7b9b3",
    "outputId": "3801c6b0-8b2f-4f0d-f9f9-12dddab469e3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    According to Gran , the company has no plans t...\n",
       "1    With the new production plant the company woul...\n",
       "2    For the last quarter of 2010 , Componenta 's n...\n",
       "3    In the third quarter of 2010 , net sales incre...\n",
       "4    Operating profit rose to EUR 13.1 mn from EUR ...\n",
       "Name: Sentence, dtype: object"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['Sentence'].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "fe6305f2-964d-486f-9d35-b886709ed5ad",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 206
    },
    "id": "fe6305f2-964d-486f-9d35-b886709ed5ad",
    "outputId": "fa4ef92d-1971-497a-c53d-527b4c789a01"
   },
   "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>Sentence</th>\n",
       "      <th>Label</th>\n",
       "      <th>Cleaned_Sentence</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>According to Gran , the company has no plans t...</td>\n",
       "      <td>neutral</td>\n",
       "      <td>according to gran the company has no plans to ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>With the new production plant the company woul...</td>\n",
       "      <td>positive</td>\n",
       "      <td>with the new production plant the company woul...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>For the last quarter of 2010 , Componenta 's n...</td>\n",
       "      <td>positive</td>\n",
       "      <td>for the last quarter of 2010 componenta s net ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>In the third quarter of 2010 , net sales incre...</td>\n",
       "      <td>positive</td>\n",
       "      <td>in the third quarter of 2010 net sales increas...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Operating profit rose to EUR 13.1 mn from EUR ...</td>\n",
       "      <td>positive</td>\n",
       "      <td>operating profit rose to eur 131 mn from eur 8...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            Sentence     Label  \\\n",
       "0  According to Gran , the company has no plans t...   neutral   \n",
       "1  With the new production plant the company woul...  positive   \n",
       "2  For the last quarter of 2010 , Componenta 's n...  positive   \n",
       "3  In the third quarter of 2010 , net sales incre...  positive   \n",
       "4  Operating profit rose to EUR 13.1 mn from EUR ...  positive   \n",
       "\n",
       "                                    Cleaned_Sentence  \n",
       "0  according to gran the company has no plans to ...  \n",
       "1  with the new production plant the company woul...  \n",
       "2  for the last quarter of 2010 componenta s net ...  \n",
       "3  in the third quarter of 2010 net sales increas...  \n",
       "4  operating profit rose to eur 131 mn from eur 8...  "
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Apply the text cleaning function to each sentence in your DataFrame.\n",
    "# Clean sentences in the DataFrame\n",
    "df['Cleaned_Sentence'] = df['Sentence'].apply(clean_text)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "d5198146-b35a-46b0-82e9-2892ed35871d",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "d5198146-b35a-46b0-82e9-2892ed35871d",
    "outputId": "54d7335b-b1aa-49ed-958d-dd5506ecd0a8"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<1492x5543 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 27277 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Use TF-IDF Vectorizer to transform cleaned sentences into numerical vectors.\n",
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "\n",
    "# Initialize and apply TF-IDF Vectorizer\n",
    "vectorizer = TfidfVectorizer()\n",
    "X = vectorizer.fit_transform(df['Cleaned_Sentence'])\n",
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "5009dede-a245-4189-ba96-5dd0b479cf0b",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "5009dede-a245-4189-ba96-5dd0b479cf0b",
    "outputId": "53bbe56b-6b9e-48b3-8c56-7c33458bdb7c"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([1, 2, 2, ..., 2, 2, 2])"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Encode the textual labels to numeric form using the LabelEncoder.\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "# Encode labels to integers\n",
    "encoder = LabelEncoder()\n",
    "y = encoder.fit_transform(df['Label'])\n",
    "y"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "2cc579eb-c01d-4511-9814-17c147620a64",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "2cc579eb-c01d-4511-9814-17c147620a64",
    "outputId": "70a46f44-d3f4-4476-c786-56a46e2f06e3"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<1193x5543 sparse matrix of type '<class 'numpy.float64'>'\n",
       "\twith 21874 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Split your data into a training set and a testing set with an 80-20 split.\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Split the dataset into training and test sets\n",
    "X_train, X_test, y_train, y_test = train_test_split(X,\n",
    "                                                    y,\n",
    "                                                    test_size=0.2,\n",
    "                                                    random_state=1502)\n",
    "X_train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "a62cbef0-ca95-4975-a066-8b9802f7c59c",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "a62cbef0-ca95-4975-a066-8b9802f7c59c",
    "outputId": "a0f9a2aa-6527-4367-e36c-964579f29a62"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "matrix([[0., 0., 0., ..., 0., 0., 0.],\n",
       "        [0., 0., 0., ..., 0., 0., 0.],\n",
       "        [0., 0., 0., ..., 0., 0., 0.],\n",
       "        ...,\n",
       "        [0., 0., 0., ..., 0., 0., 0.],\n",
       "        [0., 0., 0., ..., 0., 0., 0.],\n",
       "        [0., 0., 0., ..., 0., 0., 0.]])"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Convert the sparse matrix from vectorization into a dense matrix for model compatibility.\n",
    "# Convert sparse matrices to dense for the model input\n",
    "X_train_dense = X_train.todense()\n",
    "X_test_dense = X_test.todense()\n",
    "X_train_dense"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "8c78f29c-08c4-4c10-a77f-cb5b9d9a8f66",
   "metadata": {
    "id": "8c78f29c-08c4-4c10-a77f-cb5b9d9a8f66"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "2c2724d8-8a9c-4a48-a8c0-b0b60d658c0c",
   "metadata": {
    "id": "2c2724d8-8a9c-4a48-a8c0-b0b60d658c0c"
   },
   "source": [
    "# Task 7: Develop and Visualize a TensorFlow Model for Multi-Class Classification"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82285e7c-2407-4819-b108-411cbc8b4a5e",
   "metadata": {
    "id": "82285e7c-2407-4819-b108-411cbc8b4a5e"
   },
   "source": [
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>In this task, you will build a multi-class classification neural network model using TensorFlow and visualize its performance. You are expected to construct a Sequential model with parameters such as the number of neurons, activation functions, and output units aligned with the dataset's characteristics. You will then compile, train, and evaluate the model, followed by plotting the accuracy and loss for both training and validation data.</p><p>The specific parameters and steps for this task include:</p><ul><li><p>Construct a<strong> Sequential model </strong>with the following layers:</p></li></ul><ol><li><p>Dense layer with 64 neurons and ReLU activation.</p></li><li><p>Dropout layer with a 50% drop rate.</p></li><li><p>Dense output layer with 3 neurons (one for each class: Positive, Neutral, Negative) using the softmax activation function.</p></li></ol><ul><li><p>Compile the model using the Adam optimizer, sparse categorical cross-entropy loss, and accuracy as the metric.</p></li><li><p>Summarize the model architecture using the <code><strong>model.summary()</strong></code> method.</p></li><li><p>Train the model for 10 epochs with a batch size of 32, including validation data.</p></li><li><p>Evaluate the model's performance on the test data and print out the loss and accuracy.</p></li><li><p>Plot the training and validation accuracy, as well as the loss over epochs.</p><p><br></p></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "83206d9d-5d8c-45ca-ae3a-707b2d2f1169",
   "metadata": {
    "id": "83206d9d-5d8c-45ca-ae3a-707b2d2f1169"
   },
   "outputs": [],
   "source": [
    "# Initialize the TensorFlow Sequential model:\n",
    "# Import TensorFlow\n",
    "import tensorflow as tf\n",
    "\n",
    "# Define the Sequential model with specified layers\n",
    "model = tf.keras.Sequential([\n",
    "    tf.keras.layers.Dense(64, activation='relu', input_shape=(X_train.shape[1],)),  # 64-neuron layer\n",
    "    tf.keras.layers.Dropout(0.5),  # Dropout for regularization\n",
    "    tf.keras.layers.Dense(3, activation='softmax')  # Output layer for 3 classes\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "82ec2a2e-0bee-4ae5-88b9-f9d9272c4667",
   "metadata": {
    "id": "82ec2a2e-0bee-4ae5-88b9-f9d9272c4667"
   },
   "outputs": [],
   "source": [
    "# Compile the Sequential model:\n",
    "# Compile the model with an optimizer, loss function, and evaluation metric\n",
    "model.compile(optimizer='adam',\n",
    "              loss='sparse_categorical_crossentropy',\n",
    "              metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "839e686f-1745-48e3-a741-1781846a8ee2",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "839e686f-1745-48e3-a741-1781846a8ee2",
    "outputId": "a0f73d92-74b8-4e0b-b45a-7d3be7e64dd5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"sequential\"\n",
      "_________________________________________________________________\n",
      " Layer (type)                Output Shape              Param #   \n",
      "=================================================================\n",
      " dense (Dense)               (None, 64)                354816    \n",
      "                                                                 \n",
      " dropout (Dropout)           (None, 64)                0         \n",
      "                                                                 \n",
      " dense_1 (Dense)             (None, 3)                 195       \n",
      "                                                                 \n",
      "=================================================================\n",
      "Total params: 355,011\n",
      "Trainable params: 355,011\n",
      "Non-trainable params: 0\n",
      "_________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "# Display the model's architecture:\n",
    "# Output the model summary to understand its structure\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "9e8ba6a3-a4d2-4303-ab13-bce14e5e9610",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "9e8ba6a3-a4d2-4303-ab13-bce14e5e9610",
    "outputId": "7a39feb2-c3e3-4c1b-bbc3-2c3b34c0c56b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2024-02-04 12:10:53.577697: W tensorflow/tsl/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "38/38 [==============================] - 0s 3ms/step - loss: 1.0148 - accuracy: 0.5759 - val_loss: 0.9192 - val_accuracy: 0.5518\n",
      "Epoch 2/10\n",
      "38/38 [==============================] - 0s 2ms/step - loss: 0.8136 - accuracy: 0.6354 - val_loss: 0.7673 - val_accuracy: 0.5920\n",
      "Epoch 3/10\n",
      "38/38 [==============================] - 0s 2ms/step - loss: 0.6593 - accuracy: 0.7930 - val_loss: 0.6680 - val_accuracy: 0.7993\n",
      "Epoch 4/10\n",
      "38/38 [==============================] - 0s 1ms/step - loss: 0.5266 - accuracy: 0.8801 - val_loss: 0.5948 - val_accuracy: 0.8227\n",
      "Epoch 5/10\n",
      "38/38 [==============================] - 0s 1ms/step - loss: 0.4184 - accuracy: 0.9128 - val_loss: 0.5399 - val_accuracy: 0.8328\n",
      "Epoch 6/10\n",
      "38/38 [==============================] - 0s 1ms/step - loss: 0.3328 - accuracy: 0.9380 - val_loss: 0.5035 - val_accuracy: 0.8428\n",
      "Epoch 7/10\n",
      "38/38 [==============================] - 0s 1ms/step - loss: 0.2662 - accuracy: 0.9464 - val_loss: 0.4749 - val_accuracy: 0.8361\n",
      "Epoch 8/10\n",
      "38/38 [==============================] - 0s 2ms/step - loss: 0.2148 - accuracy: 0.9556 - val_loss: 0.4541 - val_accuracy: 0.8361\n",
      "Epoch 9/10\n",
      "38/38 [==============================] - 0s 2ms/step - loss: 0.1745 - accuracy: 0.9640 - val_loss: 0.4398 - val_accuracy: 0.8462\n",
      "Epoch 10/10\n",
      "38/38 [==============================] - 0s 1ms/step - loss: 0.1457 - accuracy: 0.9673 - val_loss: 0.4264 - val_accuracy: 0.8395\n"
     ]
    }
   ],
   "source": [
    "# Train the model with the dataset:\n",
    "# Fit the model to the training data, validating against the test data\n",
    "history = model.fit(X_train_dense, y_train,\n",
    "                    epochs=10,\n",
    "                    batch_size=32,\n",
    "                    validation_data=(X_test_dense, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "2efbff7f-73f0-43c0-b49b-3ad6b1f9caa1",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "2efbff7f-73f0-43c0-b49b-3ad6b1f9caa1",
    "outputId": "0d862d06-140f-4f7b-c5a5-2ae84ed44274"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 [==============================] - 0s 770us/step - loss: 0.4264 - accuracy: 0.8395\n",
      "Loss: 0.42643842101097107\n",
      "Accuracy: 0.8394649028778076\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the trained model's performance:\n",
    "# Evaluate the model on test dataset\n",
    "loss, accuracy = model.evaluate(X_test_dense, y_test)\n",
    "# Print loss and accuracy\n",
    "print(f'Loss: {loss}')\n",
    "print(f'Accuracy: {accuracy}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "0a0e5739-40ca-464f-b163-6a524292a7d4",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 607
    },
    "id": "0a0e5739-40ca-464f-b163-6a524292a7d4",
    "outputId": "36911946-0809-4db1-c327-f2eeb372d5fc"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1200x600 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize the training and validation metrics:\n",
    "# Import matplotlib for plotting\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# Set up the plotting environment\n",
    "plt.figure(figsize=(12,6))\n",
    "\n",
    "# Plot accuracy\n",
    "plt.subplot(1,2,1)\n",
    "plt.plot(history.history['accuracy'])\n",
    "plt.plot(history.history['val_accuracy'])\n",
    "plt.title('Model Accuracy')\n",
    "plt.ylabel('Accuracy')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Test'], loc='upper left')\n",
    "\n",
    "# Plot loss\n",
    "plt.subplot(1,2,2)\n",
    "plt.plot(history.history['loss'])\n",
    "plt.plot(history.history['val_loss'])\n",
    "plt.title('Model Loss')\n",
    "plt.ylabel('Loss')\n",
    "plt.xlabel('Epoch')\n",
    "plt.legend(['Train', 'Test'], loc='upper left')\n",
    "\n",
    "# Adjust the layout and display the plots\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "5f74a475-6f3c-4492-a591-af35211a10d0",
   "metadata": {
    "id": "5f74a475-6f3c-4492-a591-af35211a10d0"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "809e26eb-9f5e-43ce-af4c-fa860bfb6a1b",
   "metadata": {
    "id": "809e26eb-9f5e-43ce-af4c-fa860bfb6a1b"
   },
   "source": [
    "# Task 8: Optimize Neural Network Hyperparameters Using Keras Tuner"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ed0e5887-81d0-4fd3-a976-e17c1626e15d",
   "metadata": {
    "id": "ed0e5887-81d0-4fd3-a976-e17c1626e15d"
   },
   "source": [
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>This task involves using Keras Tuner to find the optimal hyperparameters for a multi-class classification neural network model. You will set up a hypermodel function that defines the model architecture and the hyperparameter space to explore. The tuner will then run multiple trials to determine the best combination of neurons in the dense layer, activation function, and dropout rate to achieve the highest validation accuracy.</p><p>The task parameters are:</p><ul><li><p>The number of units in the dense layer ranging from 64 to 512 with a step size of 64.</p></li><li><p>Activation function choices between 'relu' and 'tanh'.</p></li><li><p>Dropout rate from 0.0 to 0.5 with a step size of 0.1.</p></li><li><p>A maximum of 20 trials, with each trial executed 3 times to average out the performance variability.</p></li><li><p><br></p></li></ul></div>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "92c7bd61-d64b-437b-9eb8-0373268e1012",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "92c7bd61-d64b-437b-9eb8-0373268e1012",
    "outputId": "9d8514d9-0fbb-4515-cc57-e19230c7eabc"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: keras-tuner==1.4.5 in /Users/debdatta/anaconda3/lib/python3.10/site-packages (1.4.5)\n",
      "Requirement already satisfied: requests in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-tuner==1.4.5) (2.28.1)\n",
      "Requirement already satisfied: keras-core in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-tuner==1.4.5) (0.1.7)\n",
      "Requirement already satisfied: packaging in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-tuner==1.4.5) (22.0)\n",
      "Requirement already satisfied: kt-legacy in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-tuner==1.4.5) (1.0.5)\n",
      "Requirement already satisfied: dm-tree in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-core->keras-tuner==1.4.5) (0.1.8)\n",
      "Requirement already satisfied: namex in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-core->keras-tuner==1.4.5) (0.0.7)\n",
      "Requirement already satisfied: absl-py in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-core->keras-tuner==1.4.5) (2.1.0)\n",
      "Requirement already satisfied: numpy in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-core->keras-tuner==1.4.5) (1.23.5)\n",
      "Requirement already satisfied: h5py in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-core->keras-tuner==1.4.5) (3.7.0)\n",
      "Requirement already satisfied: rich in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from keras-core->keras-tuner==1.4.5) (13.7.0)\n",
      "Requirement already satisfied: urllib3<1.27,>=1.21.1 in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from requests->keras-tuner==1.4.5) (1.26.14)\n",
      "Requirement already satisfied: charset-normalizer<3,>=2 in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from requests->keras-tuner==1.4.5) (2.0.4)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from requests->keras-tuner==1.4.5) (2023.11.17)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from requests->keras-tuner==1.4.5) (3.4)\n",
      "Requirement already satisfied: markdown-it-py>=2.2.0 in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from rich->keras-core->keras-tuner==1.4.5) (3.0.0)\n",
      "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from rich->keras-core->keras-tuner==1.4.5) (2.17.2)\n",
      "Requirement already satisfied: mdurl~=0.1 in /Users/debdatta/anaconda3/lib/python3.10/site-packages (from markdown-it-py>=2.2.0->rich->keras-core->keras-tuner==1.4.5) (0.1.2)\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install keras-tuner==1.4.5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "e20efcca-ef44-4783-922f-aa595db21fc8",
   "metadata": {
    "id": "e20efcca-ef44-4783-922f-aa595db21fc8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Using TensorFlow backend\n"
     ]
    }
   ],
   "source": [
    "# Import the required library:\n",
    "# Import Keras Tuner\n",
    "import keras_tuner as kt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "90c46169-9ec9-4228-9b93-18409848942b",
   "metadata": {
    "id": "90c46169-9ec9-4228-9b93-18409848942b"
   },
   "outputs": [],
   "source": [
    "# Define the hypermodel function:\n",
    "\n",
    "# Function to build the model given hyperparameters\n",
    "def build_model_step1(hp):\n",
    "    model = tf.keras.Sequential()\n",
    "    # Dense layer where the number of units is a hyperparameter\n",
    "    model.add(tf.keras.layers.Dense(\n",
    "        units=hp.Int('units', min_value=64, max_value=512, step=64),\n",
    "        activation=hp.Choice('activation', values=['relu', 'tanh']),\n",
    "        input_shape=(X_train.shape[1],)\n",
    "    ))\n",
    "    # Dropout layer where the rate is a hyperparameter\n",
    "    model.add(tf.keras.layers.Dropout(\n",
    "        rate=hp.Float('dropout_rate', min_value=0.0, max_value=0.5, step=0.1)\n",
    "    ))\n",
    "    # Output layer with fixed settings\n",
    "    model.add(tf.keras.layers.Dense(3, activation='softmax'))\n",
    "    # Compile the model with default optimizer and loss function\n",
    "    model.compile(\n",
    "        optimizer=tf.keras.optimizers.Adam(),\n",
    "        loss='sparse_categorical_crossentropy',\n",
    "        metrics=['accuracy']\n",
    "    )\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "f564479a-72ad-4d4b-a8ea-72292f9da4a2",
   "metadata": {
    "id": "f564479a-72ad-4d4b-a8ea-72292f9da4a2"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    }
   ],
   "source": [
    "# Configure and run the hyperparameter search:\n",
    "\n",
    "# Initialize the random search tuner\n",
    "tuner_step1 = kt.RandomSearch(\n",
    "    build_model_step1,\n",
    "    objective='val_accuracy',\n",
    "    max_trials=20,\n",
    "    executions_per_trial=3,\n",
    "    directory='tuner_step1_directory',\n",
    "    project_name='step1_tuning'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "7aa8e280-c6d9-4532-9468-7f88402029fa",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "7aa8e280-c6d9-4532-9468-7f88402029fa",
    "outputId": "ee92cc26-2d77-454a-9429-2917cd47f1f1"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 20 Complete [00h 00m 05s]\n",
      "val_accuracy: 0.8539576331774393\n",
      "\n",
      "Best val_accuracy So Far: 0.8595317602157593\n",
      "Total elapsed time: 00h 01m 39s\n"
     ]
    }
   ],
   "source": [
    "# Start the hyperparameter tuning\n",
    "\n",
    "# Execute the search over the specified epochs and validation data\n",
    "tuner_step1.search(X_train_dense,\n",
    "                   y_train,\n",
    "                   epochs=10,\n",
    "                   validation_data=(X_test_dense,\n",
    "                                    y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "cee12552-3825-4d35-ab16-a827692ccf01",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "cee12552-3825-4d35-ab16-a827692ccf01",
    "outputId": "73b2ad2e-149f-4416-bdd6-44603b80f108"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results summary\n",
      "Results in tuner_step1_directory/step1_tuning\n",
      "Showing 10 best trials\n",
      "Objective(name=\"val_accuracy\", direction=\"max\")\n",
      "\n",
      "Trial 04 summary\n",
      "Hyperparameters:\n",
      "units: 384\n",
      "activation: relu\n",
      "dropout_rate: 0.2\n",
      "Score: 0.8595317602157593\n",
      "\n",
      "Trial 06 summary\n",
      "Hyperparameters:\n",
      "units: 384\n",
      "activation: relu\n",
      "dropout_rate: 0.1\n",
      "Score: 0.8595317602157593\n",
      "\n",
      "Trial 03 summary\n",
      "Hyperparameters:\n",
      "units: 320\n",
      "activation: relu\n",
      "dropout_rate: 0.0\n",
      "Score: 0.8584169348080953\n",
      "\n",
      "Trial 14 summary\n",
      "Hyperparameters:\n",
      "units: 384\n",
      "activation: relu\n",
      "dropout_rate: 0.30000000000000004\n",
      "Score: 0.8584169348080953\n",
      "\n",
      "Trial 00 summary\n",
      "Hyperparameters:\n",
      "units: 512\n",
      "activation: relu\n",
      "dropout_rate: 0.1\n",
      "Score: 0.8561872839927673\n",
      "\n",
      "Trial 10 summary\n",
      "Hyperparameters:\n",
      "units: 448\n",
      "activation: relu\n",
      "dropout_rate: 0.1\n",
      "Score: 0.8561872839927673\n",
      "\n",
      "Trial 13 summary\n",
      "Hyperparameters:\n",
      "units: 320\n",
      "activation: tanh\n",
      "dropout_rate: 0.30000000000000004\n",
      "Score: 0.8550724585851034\n",
      "\n",
      "Trial 12 summary\n",
      "Hyperparameters:\n",
      "units: 64\n",
      "activation: tanh\n",
      "dropout_rate: 0.2\n",
      "Score: 0.8539576331774393\n",
      "\n",
      "Trial 19 summary\n",
      "Hyperparameters:\n",
      "units: 256\n",
      "activation: relu\n",
      "dropout_rate: 0.1\n",
      "Score: 0.8539576331774393\n",
      "\n",
      "Trial 08 summary\n",
      "Hyperparameters:\n",
      "units: 512\n",
      "activation: tanh\n",
      "dropout_rate: 0.1\n",
      "Score: 0.8528428077697754\n"
     ]
    }
   ],
   "source": [
    "# Review the search results :\n",
    "\n",
    "# Output the summary of the hyperparameter tuning results\n",
    "tuner_step1.results_summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "b63acd24-b756-4ac0-87e7-173310255e92",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "b63acd24-b756-4ac0-87e7-173310255e92",
    "outputId": "6eaba231-ebc0-4905-d636-4b5879fbc951"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Units: 384\n",
      "Activation: relu\n",
      "Dropout Rate: 0.2\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Extract and print the best hyperparameters:\n",
    "\n",
    "# Get the best set of hyperparameters\n",
    "best_hps_step1 = tuner_step1.get_best_hyperparameters(num_trials=1)[0]\n",
    "\n",
    "# Print the best hyperparameters\n",
    "print(f\"\"\"\n",
    "Units: {best_hps_step1.get('units')}\n",
    "Activation: {best_hps_step1.get('activation')}\n",
    "Dropout Rate: {best_hps_step1.get('dropout_rate')}\n",
    "\"\"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "95d66021-e2ee-4eed-ac1d-6a40d671c4a7",
   "metadata": {
    "id": "95d66021-e2ee-4eed-ac1d-6a40d671c4a7"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "c41881fb-b41c-4886-a684-7548d3e64f4a",
   "metadata": {
    "id": "c41881fb-b41c-4886-a684-7548d3e64f4a"
   },
   "source": [
    "# Task 9: Refine Neural Network Model with Secondary Hyperparameter Tuning"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96c7e244-fefd-4c51-ae07-855ab1ea20d0",
   "metadata": {
    "id": "96c7e244-fefd-4c51-ae07-855ab1ea20d0"
   },
   "source": [
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>In this task, you will leverage the foundational work completed in the initial hyperparameter tuning step to further refine our Natural Language Processing (NLP) model. Your objective is to optimize the model's architecture and training configuration by introducing a second layer of hyperparameter tuning. This iterative process is designed to hone in on the best model configuration for sentiment analysis accuracy.</p><p><strong>Steps to Accomplish:</strong></p><ol><li><p><strong>Model Setup:</strong></p><ul><li><p>Utilize the <code><strong>build_model_step2</strong></code> function to construct a <code><strong>Sequential</strong></code> model.</p></li><li><p>Begin by adding a dense layer using the optimal hyperparameters (<code><strong>units</strong></code> and <code><strong>activation</strong></code>) identified from step 1 of tuning.</p></li><li><p>Incorporate a dropout layer to mitigate overfitting, again using the best dropout rate from the previous tuning step.</p></li></ul></li><li><p><strong>Hyperparameter Expansion:</strong></p><ul><li><p>Define a new hyperparameter within the dense layer to determine the ideal number of <code><strong>second_units</strong></code>. Use the provided range as your search space.</p></li><li><p>Introduce two new hyperparameters for the model: <code><strong>optimizer</strong></code> and <code><strong>learning_rate</strong></code>. You'll choose between <code><strong>adam</strong></code> and <code><strong>sgd</strong></code> for the optimizer, and among three different learning rates.</p></li></ul></li><li><p><strong>Conditional Logic Application:</strong></p><ul><li><p>Apply conditional logic to instantiate the selected optimizer with the appropriate learning rate.</p></li><li><p>Compile the model with the chosen hyperparameters, employing <code><strong>sparse_categorical_crossentropy</strong></code> as the loss function and tracking <code><strong>accuracy</strong></code> as the metric.</p></li></ul></li><li><p><strong>Tuner Initialization:</strong></p><ul><li><p>Initialize a second hyperparameter tuner (<code><strong>tuner_step2</strong></code>) using a <code><strong>RandomSearch</strong></code> strategy.</p></li><li><p>Configure this tuner with the specified search space, trial limitations, and directory setup for storing results.</p></li></ul></li><li><p><strong>Model Optimization:</strong></p><ul><li><p>Execute the hyperparameter search with the training and validation datasets provided, iterating through the specified number of epochs.</p></li><li><p>Upon completion, invoke the <code><strong>results_summary</strong></code> method to output a summary of the hyperparameter tuning results.</p><p><br></p></li></ul></li></ol></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "3a66de52-b641-40d1-a7d2-b66a867cfad1",
   "metadata": {
    "id": "3a66de52-b641-40d1-a7d2-b66a867cfad1"
   },
   "outputs": [],
   "source": [
    "# Define the function to build the refined model:\n",
    "\n",
    "def build_model_step2(hp):\n",
    "    model = tf.keras.Sequential()\n",
    "    # Add the first layer with the best hyperparameters from step 1\n",
    "    model.add(tf.keras.layers.Dense(\n",
    "        units=tuner_step1.get_best_hyperparameters()[0].get('units'),\n",
    "        activation=tuner_step1.get_best_hyperparameters()[0].get('activation'),\n",
    "        input_shape=(X_train.shape[1],)\n",
    "    ))\n",
    "    # Add the dropout layer with the best hyperparameter from step 1\n",
    "    model.add(tf.keras.layers.Dropout(\n",
    "        rate=tuner_step1.get_best_hyperparameters()[0].get('dropout_rate')\n",
    "    ))\n",
    "    # Add a second dense layer where the number of units is a new hyperparameter to tune\n",
    "    model.add(tf.keras.layers.Dense(\n",
    "        units=hp.Int('second_units', min_value=32, max_value=512, step=32),\n",
    "        activation='softmax'\n",
    "    ))\n",
    "\n",
    "    # Set optimizer and learning rate as new hyperparameters to tune\n",
    "    optimizer_selected = hp.Choice('optimizer', values=['adam', 'sgd'])\n",
    "    learning_rate_selected = hp.Choice('learning_rate', values=[1e-2, 1e-3, 1e-4])\n",
    "\n",
    "    # Conditional logic to instantiate the selected optimizer with the selected learning rate\n",
    "    if optimizer_selected == 'adam':\n",
    "        optimizer = tf.keras.optimizers.Adam(learning_rate=learning_rate_selected)\n",
    "    else:\n",
    "        optimizer = tf.keras.optimizers.SGD(learning_rate=learning_rate_selected)\n",
    "\n",
    "    # Compile the model with the selected optimizer, loss, and metric\n",
    "    model.compile(\n",
    "        optimizer=optimizer,\n",
    "        loss='sparse_categorical_crossentropy',\n",
    "        metrics=['accuracy']\n",
    "    )\n",
    "    return model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "f2ee161c-c0fa-4c32-b4d7-c5067c9163fc",
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "f2ee161c-c0fa-4c32-b4d7-c5067c9163fc",
    "outputId": "bbce00f6-a61d-4b4a-99b2-c37d62f98261"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Trial 20 Complete [00h 00m 07s]\n",
      "val_accuracy: 0.8584169348080953\n",
      "\n",
      "Best val_accuracy So Far: 0.8595317602157593\n",
      "Total elapsed time: 00h 02m 09s\n"
     ]
    }
   ],
   "source": [
    "# Initialize and run the second hyperparameter search:\n",
    "\n",
    "# Initialize the second tuner with a new search space\n",
    "tuner_step2 = kt.RandomSearch(\n",
    "    build_model_step2,\n",
    "    objective='val_accuracy',\n",
    "    max_trials=20,\n",
    "    executions_per_trial=3,\n",
    "    directory='tuner_step2_directory',\n",
    "    project_name='step2_tuning'\n",
    ")\n",
    "\n",
    "# Execute the search with the same training and validation data\n",
    "tuner_step2.search(X_train_dense, y_train, epochs=10, validation_data=(X_test_dense, y_test))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "3125b782-e969-4109-9ff4-62b3bb0a16a7",
   "metadata": {
    "id": "3125b782-e969-4109-9ff4-62b3bb0a16a7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Results summary\n",
      "Results in tuner_step2_directory/step2_tuning\n",
      "Showing 10 best trials\n",
      "Objective(name=\"val_accuracy\", direction=\"max\")\n",
      "\n",
      "Trial 08 summary\n",
      "Hyperparameters:\n",
      "second_units: 448\n",
      "optimizer: adam\n",
      "learning_rate: 0.001\n",
      "Score: 0.8595317602157593\n",
      "\n",
      "Trial 19 summary\n",
      "Hyperparameters:\n",
      "second_units: 160\n",
      "optimizer: adam\n",
      "learning_rate: 0.001\n",
      "Score: 0.8584169348080953\n",
      "\n",
      "Trial 14 summary\n",
      "Hyperparameters:\n",
      "second_units: 96\n",
      "optimizer: adam\n",
      "learning_rate: 0.001\n",
      "Score: 0.8539576331774393\n",
      "\n",
      "Trial 03 summary\n",
      "Hyperparameters:\n",
      "second_units: 320\n",
      "optimizer: adam\n",
      "learning_rate: 0.001\n",
      "Score: 0.8506131569544474\n",
      "\n",
      "Trial 05 summary\n",
      "Hyperparameters:\n",
      "second_units: 224\n",
      "optimizer: adam\n",
      "learning_rate: 0.01\n",
      "Score: 0.8494983315467834\n",
      "\n",
      "Trial 01 summary\n",
      "Hyperparameters:\n",
      "second_units: 352\n",
      "optimizer: adam\n",
      "learning_rate: 0.01\n",
      "Score: 0.8483835061391195\n",
      "\n",
      "Trial 09 summary\n",
      "Hyperparameters:\n",
      "second_units: 192\n",
      "optimizer: adam\n",
      "learning_rate: 0.01\n",
      "Score: 0.8472686608632406\n",
      "\n",
      "Trial 12 summary\n",
      "Hyperparameters:\n",
      "second_units: 448\n",
      "optimizer: adam\n",
      "learning_rate: 0.01\n",
      "Score: 0.8439242045084635\n",
      "\n",
      "Trial 04 summary\n",
      "Hyperparameters:\n",
      "second_units: 384\n",
      "optimizer: adam\n",
      "learning_rate: 0.01\n",
      "Score: 0.8439241846402487\n",
      "\n",
      "Trial 15 summary\n",
      "Hyperparameters:\n",
      "second_units: 256\n",
      "optimizer: adam\n",
      "learning_rate: 0.01\n",
      "Score: 0.8416945338249207\n"
     ]
    }
   ],
   "source": [
    "# Review the search results from the second tuning step:\n",
    "\n",
    "# Output the summary of the hyperparameter tuning results from step 2\n",
    "tuner_step2.results_summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "b851c605-3fee-4288-b55a-a9070aa1750c",
   "metadata": {
    "id": "b851c605-3fee-4288-b55a-a9070aa1750c"
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "id": "e6e5946b-2e55-4d27-8f42-9b77d42c2197",
   "metadata": {
    "id": "e6e5946b-2e55-4d27-8f42-9b77d42c2197"
   },
   "source": [
    "# Task 10: Build and Save the Final Tuned Neural Network Model"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c579fde1-54a2-46b3-be85-7e5f6319a876",
   "metadata": {
    "id": "c579fde1-54a2-46b3-be85-7e5f6319a876"
   },
   "source": [
    "<div data-purpose=\"safely-set-inner-html:rich-text-viewer:html\" class=\"lab-rich-text-viewer--lab-html-content--22XsM rt-scaffolding\"><p>In this final task, you will build a final neural network model using the best hyperparameters obtained from your comprehensive two-step hyperparameter tuning process. Once the model is configured and trained, you will save it for future use, allowing for quick deployment or further analysis without the need to retrain.</p><p>Task Steps:</p><ul><li><p><strong>Construct the final model</strong> using the best hyperparameters identified through the Keras Tuner.</p></li><li><p><strong>Train the final model</strong> using the full training dataset.</p></li><li><p><strong>Evaluate the model's performance</strong> on the test dataset to confirm its accuracy.</p></li><li><p><strong>Save the final trained model </strong>to a file.</p><p><br></p></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "94f19069-d7c6-4ccd-bd10-86c9baa03e03",
   "metadata": {
    "id": "94f19069-d7c6-4ccd-bd10-86c9baa03e03"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n",
      "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n"
     ]
    }
   ],
   "source": [
    "# Define and build the final model:\n",
    "\n",
    "def build_final_model(best_hps_step1, best_hps_step2):\n",
    "    model = tf.keras.Sequential()\n",
    "\n",
    "    # First dense layer with best units and activation from step 1\n",
    "    model.add(tf.keras.layers.Dense(\n",
    "        units=best_hps_step1.get('units'),\n",
    "        activation=best_hps_step1.get('activation'),\n",
    "        input_shape=(X_train.shape[1],)\n",
    "    ))\n",
    "\n",
    "    # Dropout layer with the best rate from step 1\n",
    "    model.add(tf.keras.layers.Dropout(best_hps_step1.get('dropout_rate')))\n",
    "\n",
    "    # Second dense layer with best units from step 2\n",
    "    model.add(tf.keras.layers.Dense(\n",
    "        units=best_hps_step2.get('second_units'),\n",
    "        activation='softmax'\n",
    "    ))\n",
    "\n",
    "    # Set the optimizer with the best type and learning rate from step 2\n",
    "    if best_hps_step2.get('optimizer') == 'adam':\n",
    "        optimizer = tf.keras.optimizers.Adam(learning_rate=best_hps_step2.get('learning_rate'))\n",
    "    else:\n",
    "        optimizer = tf.keras.optimizers.SGD(learning_rate=best_hps_step2.get('learning_rate'))\n",
    "\n",
    "    # Compile the model\n",
    "    model.compile(\n",
    "        optimizer=optimizer,\n",
    "        loss='sparse_categorical_crossentropy',\n",
    "        metrics=['accuracy']\n",
    "    )\n",
    "\n",
    "    return model\n",
    "\n",
    "# Retrieve the best hyperparameters from the first tuner\n",
    "best_hps_step1 = tuner_step1.get_best_hyperparameters()[0]\n",
    "\n",
    "# Retrieve the best hyperparameters from the second tuner\n",
    "best_hps_step2 = tuner_step2.get_best_hyperparameters()[0]\n",
    "\n",
    "# Build the final model using the best hyperparameters from both tuning steps\n",
    "final_model = build_final_model(best_hps_step1, best_hps_step2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "d15ff4c1-8a80-436b-b054-fe6ff5fe5c2e",
   "metadata": {
    "id": "d15ff4c1-8a80-436b-b054-fe6ff5fe5c2e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/10\n",
      "38/38 [==============================] - 0s 7ms/step - loss: 5.3166 - accuracy: 0.5574 - val_loss: 3.6633 - val_accuracy: 0.5485\n",
      "Epoch 2/10\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 1.5511 - accuracy: 0.6521 - val_loss: 0.7730 - val_accuracy: 0.7324\n",
      "Epoch 3/10\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.5837 - accuracy: 0.8676 - val_loss: 0.6335 - val_accuracy: 0.7726\n",
      "Epoch 4/10\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.3951 - accuracy: 0.9279 - val_loss: 0.5411 - val_accuracy: 0.8261\n",
      "Epoch 5/10\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.2572 - accuracy: 0.9514 - val_loss: 0.4823 - val_accuracy: 0.8428\n",
      "Epoch 6/10\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.1673 - accuracy: 0.9690 - val_loss: 0.4545 - val_accuracy: 0.8462\n",
      "Epoch 7/10\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.1112 - accuracy: 0.9799 - val_loss: 0.4304 - val_accuracy: 0.8462\n",
      "Epoch 8/10\n",
      "38/38 [==============================] - 0s 6ms/step - loss: 0.0728 - accuracy: 0.9866 - val_loss: 0.4208 - val_accuracy: 0.8528\n",
      "Epoch 9/10\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.0477 - accuracy: 0.9966 - val_loss: 0.4133 - val_accuracy: 0.8395\n",
      "Epoch 10/10\n",
      "38/38 [==============================] - 0s 5ms/step - loss: 0.0326 - accuracy: 0.9992 - val_loss: 0.4124 - val_accuracy: 0.8495\n"
     ]
    }
   ],
   "source": [
    "# Train the final model:\n",
    "\n",
    "# Train the final model with the training data\n",
    "history_final = final_model.fit(\n",
    "    X_train_dense,\n",
    "    y_train,\n",
    "    epochs=10,\n",
    "    batch_size=32,\n",
    "    validation_data=(X_test_dense, y_test)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "70394213-e950-4c6d-8780-5b2c905728a4",
   "metadata": {
    "id": "70394213-e950-4c6d-8780-5b2c905728a4"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10/10 [==============================] - 0s 2ms/step - loss: 0.4124 - accuracy: 0.8495\n",
      "Final Model Loss: 0.4123738706111908\n",
      "Final Model Accuracy: 0.8494983315467834\n"
     ]
    }
   ],
   "source": [
    "# Evaluate the final model's performance:\n",
    "\n",
    "# Evaluate the final model on the test data\n",
    "final_loss, final_accuracy = final_model.evaluate(X_test_dense, y_test)\n",
    "print(f'Final Model Loss: {final_loss}')\n",
    "print(f'Final Model Accuracy: {final_accuracy}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "40993594-4948-4da1-91b9-04168bf2a682",
   "metadata": {
    "id": "40993594-4948-4da1-91b9-04168bf2a682"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Final model saved to 'final_model.h5'\n"
     ]
    }
   ],
   "source": [
    "# Save the final model:\n",
    "\n",
    "# Save the final model to a file\n",
    "final_model.save('final_model.h5')\n",
    "print(\"Final model saved to 'final_model.h5'\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "0dbf3a33-1fb0-447c-a90e-ced9122ddaa8",
   "metadata": {
    "id": "0dbf3a33-1fb0-447c-a90e-ced9122ddaa8"
   },
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "colab": {
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}