{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "# Clone Assamese Tokenizer GitHub repo\n",
        "!git clone https://github.com/KashyapKishore/AssameseTokenizer.git\n",
        "\n",
        "# Add repo to Python path and import\n",
        "import sys\n",
        "sys.path.append(\"/content/AssameseTokenizer\")\n",
        "\n",
        "from assamesetokenizer import tokenize_assamese, preprocess_text\n",
        "as_tokenizer = tokenize_assamese\n",
        "\n",
        "# Core libraries\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import re\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.utils import resample, shuffle\n",
        "from sklearn.preprocessing import LabelEncoder\n",
        "from sklearn.metrics import classification_report, confusion_matrix, f1_score\n",
        "\n",
        "from tensorflow.keras.models import Sequential\n",
        "from tensorflow.keras.layers import Embedding, LSTM, Dense, SpatialDropout1D\n",
        "from tensorflow.keras.utils import to_categorical\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "from tensorflow.keras.callbacks import Callback, EarlyStopping"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YEJz0d4F46e2",
        "outputId": "169c8fa3-03cf-43b9-9727-784133858ca2"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "fatal: destination path 'AssameseTokenizer' already exists and is not an empty directory.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "g8qhPZvi5SCh",
        "outputId": "e1fe2bda-325d-47e7-8b0b-d2bf863d675e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Load the original CSV\n",
        "df = pd.read_csv('/content/drive/MyDrive/new 1.csv', header=None)\n",
        "df.columns = ['text', 'sentiment']\n",
        "df.dropna(inplace=True)\n",
        "\n",
        "# Clean text (remove non-Assamese chars, etc.)\n",
        "df['text'] = df['text'].astype(str).apply(lambda x: re.sub(r'[^\\u0980-\\u09ff\\s]', '', x))\n",
        "df['text'] = df['text'].str.replace('rt', '', regex=False)\n",
        "df.drop_duplicates(subset='text', inplace=True)\n",
        "\n",
        "from assamesetokenizer import tokenize_assamese\n",
        "\n",
        "df['tokens'] = df['text'].apply(tokenize_assamese)"
      ],
      "metadata": {
        "id": "90VqTHqtGNcH"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "df['text'] = df['text'].astype(str).apply(lambda x: re.sub(r'[^\\u0980-\\u09ff\\s]', '', x))\n",
        "df['text'] = df['text'].str.replace('rt', '', regex=False)\n",
        "df.drop_duplicates(subset='text', inplace=True)\n",
        "\n",
        "\n",
        "# Encode labels\n",
        "le = LabelEncoder()\n",
        "df['label'] = le.fit_transform(df['sentiment'])\n",
        "\n",
        "# Filter out classes with fewer than 2 samples\n",
        "class_counts = df['label'].value_counts()\n",
        "valid_classes = class_counts[class_counts >= 2].index\n",
        "df = df[df['label'].isin(valid_classes)]\n",
        "\n",
        "\n",
        "# Balance classes\n",
        "min_class_size = df['label'].value_counts().min()\n",
        "balanced_df = pd.concat([\n",
        "    resample(df[df['label'] == i], replace=True, n_samples=min_class_size, random_state=42)\n",
        "    for i in df['label'].unique()\n",
        "])\n",
        "balanced_df = balanced_df.sample(frac=1, random_state=42)"
      ],
      "metadata": {
        "id": "FkPwwpjd5kqs"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Sentiment class distribution\n",
        "sns.countplot(x='sentiment', data=df)\n",
        "plt.title(\"Class Distribution\")\n",
        "plt.show()\n",
        "\n",
        "# Sequence length\n",
        "df['seq_len'] = df['tokens'].apply(len)\n",
        "sns.histplot(df['seq_len'], bins=30)\n",
        "plt.title(\"Token Count Distribution\")\n",
        "plt.xlabel(\"Length\")\n",
        "plt.show()\n",
        "\n",
        "print(\"Max token length:\", df['seq_len'].max())\n",
        "print(\"95th percentile length:\", df['seq_len'].quantile(0.95))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 962
        },
        "id": "m8oS15QTF2Up",
        "outputId": "d4aeeb66-f8fc-430a-fd5b-43d62fb9cf99"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAHHCAYAAACiOWx7AAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjAsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvlHJYcgAAAAlwSFlzAAAPYQAAD2EBqD+naQAARolJREFUeJzt3XtUVXX+//HXAT0HRA9eAUkES0fFUPMSonbRyKPhTE7aaDlqXr86qCnlrfGrZpmNfU1NMytnpCb9ptZkJokS3iZlvFDeL5lZWAqaCUctAWH//ujL/nmEdIvkQXs+1jprefbnvT/7vffaC17us8/GZhiGIQAAAFyRj7cbAAAAuBkQmgAAACwgNAEAAFhAaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAlEpERISeeOIJb7dx3aZMmSKbzXZDtnX//ffr/vvvN99v2LBBNptN77333g3Z/hNPPKGIiIgbsi3gVkRoAuDhyJEj+q//+i/dfvvt8vPzk9PpVLt27TRnzhz99NNP3m7vihITE2Wz2cyXn5+fQkND5XK59Morr+js2bNlsp3jx49rypQp2rlzZ5nMV5bKc2/Aza6CtxsAUH4kJSXp0UcflcPhUN++fXXnnXcqLy9Pn376qcaMGaN9+/bpjTfe8HabVzV16lTVq1dP+fn5yszM1IYNGzRq1Ci9/PLLWrlypZo2bWrWTpw4UePHj7+m+Y8fP65nn31WERERat68ueX11q5de03bKY0r9fbmm2+qsLDwV+8BuFURmgBIko4ePapevXopPDxc69atU+3atc2x+Ph4ffnll0pKSvJih9Z16dJFrVq1Mt9PmDBB69atU9euXfWHP/xBBw4ckL+/vySpQoUKqlDh1/1R+OOPP6pSpUqy2+2/6naupmLFil7dPnCz4+M5AJKkGTNm6Ny5c/r73//uEZiK1K9fX08++eQvrv/DDz/o6aefVlRUlCpXriyn06kuXbpo165dxWrnzp2rJk2aqFKlSqpWrZpatWqlJUuWmONnz57VqFGjFBERIYfDoaCgID344IP67LPPSr1/HTt21H//93/rm2++0TvvvGMuL+meppSUFLVv315Vq1ZV5cqV1bBhQz3zzDOSfr4PqXXr1pKk/v37mx8FJiYmSvr5vqU777xT6enpuvfee1WpUiVz3cvvaSpSUFCgZ555RiEhIQoICNAf/vAHHTt2zKPml+4hu3TOq/VW0j1N58+f11NPPaWwsDA5HA41bNhQ//M//yPDMDzqbDabhg8frhUrVujOO++Uw+FQkyZNlJycXPIBB25BXGkCIEn66KOPdPvtt6tt27alWv+rr77SihUr9Oijj6pevXrKysrS66+/rvvuu0/79+9XaGiopJ8/Iho5cqR69OihJ598UhcuXNDu3bu1detWPf7445KkoUOH6r333tPw4cMVGRmp06dP69NPP9WBAwfUokWLUu9jnz599Mwzz2jt2rUaPHhwiTX79u1T165d1bRpU02dOlUOh0NffvmlNm/eLElq3Lixpk6dqkmTJmnIkCG65557JMnjuJ0+fVpdunRRr1699Oc//1nBwcFX7GvatGmy2WwaN26cTp48qdmzZys2NlY7d+40r4hZYaW3SxmGoT/84Q9av369Bg4cqObNm2vNmjUaM2aMvvvuO82aNcuj/tNPP9W//vUv/eUvf1GVKlX0yiuvqHv37srIyFCNGjUs9wnctAwAv3k5OTmGJOPhhx+2vE54eLjRr18/8/2FCxeMgoICj5qjR48aDofDmDp1qrns4YcfNpo0aXLFuQMDA434+HjLvRRZtGiRIcnYvn37Fee+6667zPeTJ082Lv1ROGvWLEOScerUqV+cY/v27YYkY9GiRcXG7rvvPkOSsWDBghLH7rvvPvP9+vXrDUnGbbfdZrjdbnP5smXLDEnGnDlzzGWXH+9fmvNKvfXr188IDw83369YscKQZDz//PMedT169DBsNpvx5ZdfmsskGXa73WPZrl27DEnG3Llzi20LuBXx8RwAud1uSVKVKlVKPYfD4ZCPz88/UgoKCnT69Gnzo61LP1arWrWqvv32W23fvv0X56pataq2bt2q48ePl7qfX1K5cuUrfouuatWqkqQPP/yw1DdNOxwO9e/f33J93759PY59jx49VLt2bX388cel2r5VH3/8sXx9fTVy5EiP5U899ZQMw9Dq1as9lsfGxuqOO+4w3zdt2lROp1NfffXVr9onUF4QmgDI6XRK0nV9Jb+wsFCzZs1SgwYN5HA4VLNmTdWqVUu7d+9WTk6OWTdu3DhVrlxZd999txo0aKD4+Hjzo68iM2bM0N69exUWFqa7775bU6ZMKbNfzOfOnbtiOOzZs6fatWunQYMGKTg4WL169dKyZcuuKUDddttt13TTd4MGDTze22w21a9fX19//bXlOUrjm2++UWhoaLHj0bhxY3P8UnXr1i02R7Vq1XTmzJlfr0mgHCE0AZDT6VRoaKj27t1b6jleeOEFJSQk6N5779U777yjNWvWKCUlRU2aNPEIHI0bN9ahQ4f07rvvqn379nr//ffVvn17TZ482az505/+pK+++kpz585VaGioXnrpJTVp0qTYlY9r9e233yonJ0f169f/xRp/f39t2rRJn3zyifr06aPdu3erZ8+eevDBB1VQUGBpO9dyH5JVv/QATqs9lQVfX98SlxuX3TQO3KoITQAkSV27dtWRI0eUlpZWqvXfe+89dejQQX//+9/Vq1cvderUSbGxscrOzi5WGxAQoJ49e2rRokXKyMhQXFycpk2bpgsXLpg1tWvX1l/+8hetWLFCR48eVY0aNTRt2rTS7p4k6Z///KckyeVyXbHOx8dHDzzwgF5++WXt379f06ZN07p167R+/XpJvxxgSuvw4cMe7w3D0JdffunxTbdq1aqVeCwvvxp0Lb2Fh4fr+PHjxa4wHjx40BwH8P8RmgBIksaOHauAgAANGjRIWVlZxcaPHDmiOXPm/OL6vr6+xa44LF++XN99953HstOnT3u8t9vtioyMlGEYys/PV0FBgcfHeZIUFBSk0NBQ5ebmXutumdatW6fnnntO9erVU+/evX+x7ocffii2rOghkUXbDwgIkKQSQ0xpvP322x7B5b333tOJEyfUpUsXc9kdd9yh//znP8rLyzOXrVq1qtijCa6lt4ceekgFBQWaN2+ex/JZs2bJZrN5bB8AjxwA8H/uuOMOLVmyRD179lTjxo09ngi+ZcsWLV++/Ip/a65r166aOnWq+vfvr7Zt22rPnj1avHixbr/9do+6Tp06KSQkRO3atVNwcLAOHDigefPmKS4uTlWqVFF2drbq1KmjHj16qFmzZqpcubI++eQTbd++XTNnzrS0L6tXr9bBgwd18eJFZWVlad26dUpJSVF4eLhWrlwpPz+/X1x36tSp2rRpk+Li4hQeHq6TJ09q/vz5qlOnjtq3b28eq6pVq2rBggWqUqWKAgICFB0drXr16lnq73LVq1dX+/bt1b9/f2VlZWn27NmqX7++x2MRBg0apPfee0+dO3fWn/70Jx05ckTvvPOOx43Z19rb73//e3Xo0EF//etf9fXXX6tZs2Zau3atPvzwQ40aNarY3MBvnle/uweg3Pniiy+MwYMHGxEREYbdbjeqVKlitGvXzpg7d65x4cIFs66kRw489dRTRu3atQ1/f3+jXbt2RlpaWrGvxL/++uvGvffea9SoUcNwOBzGHXfcYYwZM8bIyckxDMMwcnNzjTFjxhjNmjUzqlSpYgQEBBjNmjUz5s+ff9Xeix45UPSy2+1GSEiI8eCDDxpz5szx+Fp/kcsfOZCammo8/PDDRmhoqGG3243Q0FDjscceM7744guP9T788EMjMjLSqFChgsdX/O+7775ffKTCLz1y4H//93+NCRMmGEFBQYa/v78RFxdnfPPNN8XWnzlzpnHbbbcZDofDaNeunbFjx45ic16pt8sfOWAYhnH27Flj9OjRRmhoqFGxYkWjQYMGxksvvWQUFhZ61Ekq8TEQv/QoBOBWZDMM7uADAAC4Gu5pAgAAsIDQBAAAYAGhCQAAwAJCEwAAgAWEJgAAAAsITQAAABbwcMsyUlhYqOPHj6tKlSpl/icWAADAr8MwDJ09e1ahoaHy8bnytSRCUxk5fvy4wsLCvN0GAAAohWPHjqlOnTpXrCE0lZEqVapI+vmgO51OL3cDAACscLvdCgsLM3+PXwmhqYwUfSTndDoJTQAA3GSs3FrDjeAAAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAACwhNAAAAFhCaAAAALCA0AQAAWEBoAgAAsIDQBAAAYAGhCQAAwAJCEwAAgAUVvN0AgPKt5Zi3vd0CypH0l/p6uwXAa7jSBAAAYAGhCQAAwAJCEwAAgAVeDU0RERGy2WzFXvHx8ZKkCxcuKD4+XjVq1FDlypXVvXt3ZWVlecyRkZGhuLg4VapUSUFBQRozZowuXrzoUbNhwwa1aNFCDodD9evXV2JiYrFeXn31VUVERMjPz0/R0dHatm3br7bfAADg5uPV0LR9+3adOHHCfKWkpEiSHn30UUnS6NGj9dFHH2n58uXauHGjjh8/rkceecRcv6CgQHFxccrLy9OWLVv01ltvKTExUZMmTTJrjh49qri4OHXo0EE7d+7UqFGjNGjQIK1Zs8asWbp0qRISEjR58mR99tlnatasmVwul06ePHmDjgQAACjvbIZhGN5uosioUaO0atUqHT58WG63W7Vq1dKSJUvUo0cPSdLBgwfVuHFjpaWlqU2bNlq9erW6du2q48ePKzg4WJK0YMECjRs3TqdOnZLdbte4ceOUlJSkvXv3mtvp1auXsrOzlZycLEmKjo5W69atNW/ePElSYWGhwsLCNGLECI0fP95S7263W4GBgcrJyZHT6SzLwwJ4Fd+ew6X49hxuNdfy+7vc3NOUl5end955RwMGDJDNZlN6erry8/MVGxtr1jRq1Eh169ZVWlqaJCktLU1RUVFmYJIkl8slt9utffv2mTWXzlFUUzRHXl6e0tPTPWp8fHwUGxtr1pQkNzdXbrfb4wUAAG5d5SY0rVixQtnZ2XriiSckSZmZmbLb7apatapHXXBwsDIzM82aSwNT0XjR2JVq3G63fvrpJ33//fcqKCgosaZojpJMnz5dgYGB5issLOya9xkAANw8yk1o+vvf/64uXbooNDTU261YMmHCBOXk5JivY8eOebslAADwKyoXTwT/5ptv9Mknn+hf//qXuSwkJER5eXnKzs72uNqUlZWlkJAQs+byb7kVfbvu0prLv3GXlZUlp9Mpf39/+fr6ytfXt8SaojlK4nA45HA4rn1nAQDATalcXGlatGiRgoKCFBcXZy5r2bKlKlasqNTUVHPZoUOHlJGRoZiYGElSTEyM9uzZ4/Ett5SUFDmdTkVGRpo1l85RVFM0h91uV8uWLT1qCgsLlZqaatYAAAB4/UpTYWGhFi1apH79+qlChf/fTmBgoAYOHKiEhARVr15dTqdTI0aMUExMjNq0aSNJ6tSpkyIjI9WnTx/NmDFDmZmZmjhxouLj482rQEOHDtW8efM0duxYDRgwQOvWrdOyZcuUlJRkbishIUH9+vVTq1atdPfdd2v27Nk6f/68+vfvf2MPBgAAKLe8Hpo++eQTZWRkaMCAAcXGZs2aJR8fH3Xv3l25ublyuVyaP3++Oe7r66tVq1Zp2LBhiomJUUBAgPr166epU6eaNfXq1VNSUpJGjx6tOXPmqE6dOlq4cKFcLpdZ07NnT506dUqTJk1SZmammjdvruTk5GI3hwMAgN+ucvWcppsZz2nCrYrnNOFSPKcJt5qb8jlNAAAA5RmhCQAAwAJCEwAAgAWEJgAAAAsITQAAABYQmgAAACwgNAEAAFhAaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAACwhNAAAAFhCaAAAALCA0AQAAWEBoAgAAsIDQBAAAYAGhCQAAwAJCEwAAgAWEJgAAAAsITQAAABYQmgAAACwgNAEAAFhAaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAs8Hpo+u677/TnP/9ZNWrUkL+/v6KiorRjxw5z3DAMTZo0SbVr15a/v79iY2N1+PBhjzl++OEH9e7dW06nU1WrVtXAgQN17tw5j5rdu3frnnvukZ+fn8LCwjRjxoxivSxfvlyNGjWSn5+foqKi9PHHH/86Ow0AAG46Xg1NZ86cUbt27VSxYkWtXr1a+/fv18yZM1WtWjWzZsaMGXrllVe0YMECbd26VQEBAXK5XLpw4YJZ07t3b+3bt08pKSlatWqVNm3apCFDhpjjbrdbnTp1Unh4uNLT0/XSSy9pypQpeuONN8yaLVu26LHHHtPAgQP1+eefq1u3burWrZv27t17Yw4GAAAo12yGYRje2vj48eO1efNm/fvf/y5x3DAMhYaG6qmnntLTTz8tScrJyVFwcLASExPVq1cvHThwQJGRkdq+fbtatWolSUpOTtZDDz2kb7/9VqGhoXrttdf017/+VZmZmbLb7ea2V6xYoYMHD0qSevbsqfPnz2vVqlXm9tu0aaPmzZtrwYIFV90Xt9utwMBA5eTkyOl0XtdxAcqTlmPe9nYLKEfSX+rr7RaAMnUtv7+9eqVp5cqVatWqlR599FEFBQXprrvu0ptvvmmOHz16VJmZmYqNjTWXBQYGKjo6WmlpaZKktLQ0Va1a1QxMkhQbGysfHx9t3brVrLn33nvNwCRJLpdLhw4d0pkzZ8yaS7dTVFO0ncvl5ubK7XZ7vAAAwK3Lq6Hpq6++0muvvaYGDRpozZo1GjZsmEaOHKm33npLkpSZmSlJCg4O9lgvODjYHMvMzFRQUJDHeIUKFVS9enWPmpLmuHQbv1RTNH656dOnKzAw0HyFhYVd8/4DAICbh1dDU2FhoVq0aKEXXnhBd911l4YMGaLBgwdb+jjM2yZMmKCcnBzzdezYMW+3BAAAfkVeDU21a9dWZGSkx7LGjRsrIyNDkhQSEiJJysrK8qjJysoyx0JCQnTy5EmP8YsXL+qHH37wqClpjku38Us1ReOXczgccjqdHi8AAHDr8mpoateunQ4dOuSx7IsvvlB4eLgkqV69egoJCVFqaqo57na7tXXrVsXExEiSYmJilJ2drfT0dLNm3bp1KiwsVHR0tFmzadMm5efnmzUpKSlq2LCh+U29mJgYj+0U1RRtBwAA/LZ5NTSNHj1a//nPf/TCCy/oyy+/1JIlS/TGG28oPj5ekmSz2TRq1Cg9//zzWrlypfbs2aO+ffsqNDRU3bp1k/TzlanOnTtr8ODB2rZtmzZv3qzhw4erV69eCg0NlSQ9/vjjstvtGjhwoPbt26elS5dqzpw5SkhIMHt58sknlZycrJkzZ+rgwYOaMmWKduzYoeHDh9/w4wIAAMqfCt7ceOvWrfXBBx9owoQJmjp1qurVq6fZs2erd+/eZs3YsWN1/vx5DRkyRNnZ2Wrfvr2Sk5Pl5+dn1ixevFjDhw/XAw88IB8fH3Xv3l2vvPKKOR4YGKi1a9cqPj5eLVu2VM2aNTVp0iSPZzm1bdtWS5Ys0cSJE/XMM8+oQYMGWrFihe68884bczAAAEC55tXnNN1KeE4TblU8pwmX4jlNuNXcNM9pAgAAuFkQmgAAACwgNAEAAFhAaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAACwhNAAAAFhCaAAAALCA0AQAAWEBoAgAAsIDQBAAAYAGhCQAAwAJCEwAAgAWEJgAAAAsITQAAABYQmgAAACwgNAEAAFhAaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAq+GpilTpshms3m8GjVqZI5fuHBB8fHxqlGjhipXrqzu3bsrKyvLY46MjAzFxcWpUqVKCgoK0pgxY3Tx4kWPmg0bNqhFixZyOByqX7++EhMTi/Xy6quvKiIiQn5+foqOjta2bdt+lX0GAAA3J69faWrSpIlOnDhhvj799FNzbPTo0froo4+0fPlybdy4UcePH9cjjzxijhcUFCguLk55eXnasmWL3nrrLSUmJmrSpElmzdGjRxUXF6cOHTpo586dGjVqlAYNGqQ1a9aYNUuXLlVCQoImT56szz77TM2aNZPL5dLJkydvzEEAAADlns0wDMNbG58yZYpWrFihnTt3FhvLyclRrVq1tGTJEvXo0UOSdPDgQTVu3FhpaWlq06aNVq9era5du+r48eMKDg6WJC1YsEDjxo3TqVOnZLfbNW7cOCUlJWnv3r3m3L169VJ2draSk5MlSdHR0WrdurXmzZsnSSosLFRYWJhGjBih8ePHW9oXt9utwMBA5eTkyOl0Xs9hAcqVlmPe9nYLKEfSX+rr7RaAMnUtv7+9fqXp8OHDCg0N1e23367evXsrIyNDkpSenq78/HzFxsaatY0aNVLdunWVlpYmSUpLS1NUVJQZmCTJ5XLJ7XZr3759Zs2lcxTVFM2Rl5en9PR0jxofHx/FxsaaNSXJzc2V2+32eAEAgFuXV0NTdHS0EhMTlZycrNdee01Hjx7VPffco7NnzyozM1N2u11Vq1b1WCc4OFiZmZmSpMzMTI/AVDReNHalGrfbrZ9++knff/+9CgoKSqwpmqMk06dPV2BgoPkKCwsr1TEAAAA3hwre3HiXLl3Mfzdt2lTR0dEKDw/XsmXL5O/v78XOrm7ChAlKSEgw37vdboITAAC3MK9/PHepqlWr6ne/+52+/PJLhYSEKC8vT9nZ2R41WVlZCgkJkSSFhIQU+zZd0fur1TidTvn7+6tmzZry9fUtsaZojpI4HA45nU6PFwAAuHWVq9B07tw5HTlyRLVr11bLli1VsWJFpaammuOHDh1SRkaGYmJiJEkxMTHas2ePx7fcUlJS5HQ6FRkZadZcOkdRTdEcdrtdLVu29KgpLCxUamqqWQMAAODV0PT0009r48aN+vrrr7Vlyxb98Y9/lK+vrx577DEFBgZq4MCBSkhI0Pr165Wenq7+/fsrJiZGbdq0kSR16tRJkZGR6tOnj3bt2qU1a9Zo4sSJio+Pl8PhkCQNHTpUX331lcaOHauDBw9q/vz5WrZsmUaPHm32kZCQoDfffFNvvfWWDhw4oGHDhun8+fPq37+/V44LAAAof7x6T9O3336rxx57TKdPn1atWrXUvn17/ec//1GtWrUkSbNmzZKPj4+6d++u3NxcuVwuzZ8/31zf19dXq1at0rBhwxQTE6OAgAD169dPU6dONWvq1aunpKQkjR49WnPmzFGdOnW0cOFCuVwus6Znz546deqUJk2apMzMTDVv3lzJycnFbg4HAAC/XV59TtOthOc04VbFc5pwKZ7ThFvNTfWcJgAAgJsBoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAACwhNAAAAFhCaAAAALCA0AQAAWEBoAgAAsIDQBAAAYAGhCQAAwAJCEwAAgAWEJgAAAAsITQAAABYQmgAAACwgNAEAAFhAaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAACwhNAAAAFpQqNHXs2FHZ2dnFlrvdbnXs2PF6ewIAACh3ShWaNmzYoLy8vGLLL1y4oH//+9+lauTFF1+UzWbTqFGjPOaLj49XjRo1VLlyZXXv3l1ZWVke62VkZCguLk6VKlVSUFCQxowZo4sXLxbrt0WLFnI4HKpfv74SExOLbf/VV19VRESE/Pz8FB0drW3btpVqPwAAwK2pwrUU79692/z3/v37lZmZab4vKChQcnKybrvttmtuYvv27Xr99dfVtGlTj+WjR49WUlKSli9frsDAQA0fPlyPPPKINm/ebG4zLi5OISEh2rJli06cOKG+ffuqYsWKeuGFFyRJR48eVVxcnIYOHarFixcrNTVVgwYNUu3ateVyuSRJS5cuVUJCghYsWKDo6GjNnj1bLpdLhw4dUlBQ0DXvDwAAuPXYDMMwrBb7+PjIZrNJkkpazd/fX3PnztWAAQMsN3Du3Dm1aNFC8+fP1/PPP6/mzZtr9uzZysnJUa1atbRkyRL16NFDknTw4EE1btxYaWlpatOmjVavXq2uXbvq+PHjCg4OliQtWLBA48aN06lTp2S32zVu3DglJSVp79695jZ79eql7OxsJScnS5Kio6PVunVrzZs3T5JUWFiosLAwjRgxQuPHj7e0H263W4GBgcrJyZHT6bS8/0B513LM295uAeVI+kt9vd0CUKau5ff3NX08d/ToUR05ckSGYWjbtm06evSo+fruu+/kdruvKTBJUnx8vOLi4hQbG+uxPD09Xfn5+R7LGzVqpLp16yotLU2SlJaWpqioKDMwSZLL5ZLb7da+ffvMmsvndrlc5hx5eXlKT0/3qPHx8VFsbKxZU5Lc3Fy53W6PFwAAuHVd08dz4eHhkn6+ElMW3n33XX322Wfavn17sbHMzEzZ7XZVrVrVY3lwcLD5sWBmZqZHYCoaLxq7Uo3b7dZPP/2kM2fOqKCgoMSagwcP/mLv06dP17PPPmttRwEAwE3vmkLTpQ4fPqz169fr5MmTxULUpEmTrrr+sWPH9OSTTyolJUV+fn6lbcNrJkyYoISEBPO92+1WWFiYFzsCAAC/plKFpjfffFPDhg1TzZo1FRISYt7nJEk2m81SaEpPT9fJkyfVokULc1lBQYE2bdqkefPmac2aNcrLy1N2drbH1aasrCyFhIRIkkJCQop9y63o23WX1lz+jbusrCw5nU75+/vL19dXvr6+JdYUzVESh8Mhh8Nx1f0EAAC3hlI9cuD555/XtGnTlJmZqZ07d+rzzz83X5999pmlOR544AHt2bNHO3fuNF+tWrVS7969zX9XrFhRqamp5jqHDh1SRkaGYmJiJEkxMTHas2ePTp48adakpKTI6XQqMjLSrLl0jqKaojnsdrtatmzpUVNYWKjU1FSzBgAAoFRXms6cOaNHH330ujZcpUoV3XnnnR7LAgICVKNGDXP5wIEDlZCQoOrVq8vpdGrEiBGKiYlRmzZtJEmdOnVSZGSk+vTpoxkzZigzM1MTJ05UfHy8eRVo6NChmjdvnsaOHasBAwZo3bp1WrZsmZKSksztJiQkqF+/fmrVqpXuvvtuzZ49W+fPn1f//v2vax8BAMCto1Sh6dFHH9XatWs1dOjQsu7Hw6xZs+Tj46Pu3bsrNzdXLpdL8+fPN8d9fX21atUqDRs2TDExMQoICFC/fv00depUs6ZevXpKSkrS6NGjNWfOHNWpU0cLFy40n9EkST179tSpU6c0adIkZWZmqnnz5kpOTi52czgAAPjtuqbnNBWZPn26Xn75ZcXFxSkqKkoVK1b0GB85cmSZNXiz4DlNuFXxnCZciuc04VZzLb+/S3Wl6Y033lDlypW1ceNGbdy40WPMZrP9JkMTAAC4tZUqNB09erSs+wAAACjXSvXtOQAAgN+aUl1putqfSvnHP/5RqmYAAADKq1I/cuBS+fn52rt3r7Kzs9WxY8cyaQwAAKA8KVVo+uCDD4otKyws1LBhw3THHXdcd1MAAADlTZnd0+Tj46OEhATNmjWrrKYEAAAoN8r0RvAjR47o4sWLZTklAABAuVCqj+cSEhI83huGoRMnTigpKUn9+vUrk8YAAADKk1KFps8//9zjvY+Pj2rVqqWZM2de9Zt1AAAAN6NShab169eXdR8AAADlWqlCU5FTp07p0KFDkqSGDRuqVq1aZdIUAABAeVOqG8HPnz+vAQMGqHbt2rr33nt17733KjQ0VAMHDtSPP/5Y1j0CAAB4XalCU0JCgjZu3KiPPvpI2dnZys7O1ocffqiNGzfqqaeeKuseAQAAvK5UH8+9//77eu+993T//febyx566CH5+/vrT3/6k1577bWy6g8AAKBcKNWVph9//FHBwcHFlgcFBfHxHAAAuCWVKjTFxMRo8uTJunDhgrnsp59+0rPPPquYmJgyaw4AAKC8KNXHc7Nnz1bnzp1Vp04dNWvWTJK0a9cuORwOrV27tkwbBAAAKA9KFZqioqJ0+PBhLV68WAcPHpQkPfbYY+rdu7f8/f3LtEEAAIDyoFShafr06QoODtbgwYM9lv/jH//QqVOnNG7cuDJpDgAAoLwo1T1Nr7/+uho1alRseZMmTbRgwYLrbgoAAKC8KVVoyszMVO3atYstr1Wrlk6cOHHdTQEAAJQ3pQpNYWFh2rx5c7HlmzdvVmho6HU3BQAAUN6U6p6mwYMHa9SoUcrPz1fHjh0lSampqRo7dixPBAcAALekUoWmMWPG6PTp0/rLX/6ivLw8SZKfn5/GjRunCRMmlGmDAAAA5UGpQpPNZtPf/vY3/fd//7cOHDggf39/NWjQQA6Ho6z7AwAAKBdKFZqKVK5cWa1bty6rXgAAAMqtUt0IDgAA8FtDaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsuK6HWwIAcKO1HPO2t1tAOZP+Ut8bsh2uNAEAAFhAaAIAALDAq6HptddeU9OmTeV0OuV0OhUTE6PVq1eb4xcuXFB8fLxq1KihypUrq3v37srKyvKYIyMjQ3FxcapUqZKCgoI0ZswYXbx40aNmw4YNatGihRwOh+rXr6/ExMRivbz66quKiIiQn5+foqOjtW3btl9lnwEAwM3Jq6GpTp06evHFF5Wenq4dO3aoY8eOevjhh7Vv3z5J0ujRo/XRRx9p+fLl2rhxo44fP65HHnnEXL+goEBxcXHKy8vTli1b9NZbbykxMVGTJk0ya44ePaq4uDh16NBBO3fu1KhRozRo0CCtWbPGrFm6dKkSEhI0efJkffbZZ2rWrJlcLpdOnjx54w4GAAAo12yGYRjebuJS1atX10svvaQePXqoVq1aWrJkiXr06CFJOnjwoBo3bqy0tDS1adNGq1evVteuXXX8+HEFBwdLkhYsWKBx48bp1KlTstvtGjdunJKSkrR3715zG7169VJ2draSk5MlSdHR0WrdurXmzZsnSSosLFRYWJhGjBih8ePHW+rb7XYrMDBQOTk5cjqdZXlIAK/ipltc6kbdcHslnJO43PWcl9fy+7vc3NNUUFCgd999V+fPn1dMTIzS09OVn5+v2NhYs6ZRo0aqW7eu0tLSJElpaWmKiooyA5MkuVwuud1u82pVWlqaxxxFNUVz5OXlKT093aPGx8dHsbGxZk1JcnNz5Xa7PV4AAODW5fXQtGfPHlWuXFkOh0NDhw7VBx98oMjISGVmZsput6tq1aoe9cHBwcrMzJQkZWZmegSmovGisSvVuN1u/fTTT/r+++9VUFBQYk3RHCWZPn26AgMDzVdYWFip9h8AANwcvB6aGjZsqJ07d2rr1q0aNmyY+vXrp/3793u7rauaMGGCcnJyzNexY8e83RIAAPgVef3hlna7XfXr15cktWzZUtu3b9ecOXPUs2dP5eXlKTs72+NqU1ZWlkJCQiRJISEhxb7lVvTtuktrLv/GXVZWlpxOp/z9/eXr6ytfX98Sa4rmKInD4ZDD4SjdTgMAgJuO1680Xa6wsFC5ublq2bKlKlasqNTUVHPs0KFDysjIUExMjCQpJiZGe/bs8fiWW0pKipxOpyIjI82aS+coqimaw263q2XLlh41hYWFSk1NNWsAAAC8eqVpwoQJ6tKli+rWrauzZ89qyZIl2rBhg9asWaPAwEANHDhQCQkJql69upxOp0aMGKGYmBi1adNGktSpUydFRkaqT58+mjFjhjIzMzVx4kTFx8ebV4GGDh2qefPmaezYsRowYIDWrVunZcuWKSkpyewjISFB/fr1U6tWrXT33Xdr9uzZOn/+vPr37++V4wIAAMofr4amkydPqm/fvjpx4oQCAwPVtGlTrVmzRg8++KAkadasWfLx8VH37t2Vm5srl8ul+fPnm+v7+vpq1apVGjZsmGJiYhQQEKB+/fpp6tSpZk29evWUlJSk0aNHa86cOapTp44WLlwol8tl1vTs2VOnTp3SpEmTlJmZqebNmys5ObnYzeEAAOC3q9w9p+lmxXOacKvimTi4FM9pQnn0m3tOEwAAQHlGaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAACwhNAAAAFhCaAAAALCA0AQAAWEBoAgAAsIDQBAAAYAGhCQAAwAJCEwAAgAWEJgAAAAsITQAAABYQmgAAACwgNAEAAFhAaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAAC7wamqZPn67WrVurSpUqCgoKUrdu3XTo0CGPmgsXLig+Pl41atRQ5cqV1b17d2VlZXnUZGRkKC4uTpUqVVJQUJDGjBmjixcvetRs2LBBLVq0kMPhUP369ZWYmFisn1dffVURERHy8/NTdHS0tm3bVub7DAAAbk5eDU0bN25UfHy8/vOf/yglJUX5+fnq1KmTzp8/b9aMHj1aH330kZYvX66NGzfq+PHjeuSRR8zxgoICxcXFKS8vT1u2bNFbb72lxMRETZo0yaw5evSo4uLi1KFDB+3cuVOjRo3SoEGDtGbNGrNm6dKlSkhI0OTJk/XZZ5+pWbNmcrlcOnny5I05GAAAoFyzGYZheLuJIqdOnVJQUJA2btyoe++9Vzk5OapVq5aWLFmiHj16SJIOHjyoxo0bKy0tTW3atNHq1avVtWtXHT9+XMHBwZKkBQsWaNy4cTp16pTsdrvGjRunpKQk7d2719xWr169lJ2dreTkZElSdHS0WrdurXnz5kmSCgsLFRYWphEjRmj8+PFX7d3tdiswMFA5OTlyOp1lfWgAr2k55m1vt4ByJP2lvt5ugXMSxVzPeXktv7/L1T1NOTk5kqTq1atLktLT05Wfn6/Y2FizplGjRqpbt67S0tIkSWlpaYqKijIDkyS5XC653W7t27fPrLl0jqKaojny8vKUnp7uUePj46PY2FizBgAA/LZV8HYDRQoLCzVq1Ci1a9dOd955pyQpMzNTdrtdVatW9agNDg5WZmamWXNpYCoaLxq7Uo3b7dZPP/2kM2fOqKCgoMSagwcPlthvbm6ucnNzzfdut/sa9xgAANxMys2Vpvj4eO3du1fvvvuut1uxZPr06QoMDDRfYWFh3m4JAAD8ispFaBo+fLhWrVql9evXq06dOubykJAQ5eXlKTs726M+KytLISEhZs3l36Yren+1GqfTKX9/f9WsWVO+vr4l1hTNcbkJEyYoJyfHfB07duzadxwAANw0vBqaDMPQ8OHD9cEHH2jdunWqV6+ex3jLli1VsWJFpaammssOHTqkjIwMxcTESJJiYmK0Z88ej2+5paSkyOl0KjIy0qy5dI6imqI57Ha7WrZs6VFTWFio1NRUs+ZyDodDTqfT4wUAAG5dXr2nKT4+XkuWLNGHH36oKlWqmPcgBQYGyt/fX4GBgRo4cKASEhJUvXp1OZ1OjRgxQjExMWrTpo0kqVOnToqMjFSfPn00Y8YMZWZmauLEiYqPj5fD4ZAkDR06VPPmzdPYsWM1YMAArVu3TsuWLVNSUpLZS0JCgvr166dWrVrp7rvv1uzZs3X+/Hn179//xh8YAABQ7ng1NL322muSpPvvv99j+aJFi/TEE09IkmbNmiUfHx91795dubm5crlcmj9/vlnr6+urVatWadiwYYqJiVFAQID69eunqVOnmjX16tVTUlKSRo8erTlz5qhOnTpauHChXC6XWdOzZ0+dOnVKkyZNUmZmppo3b67k5ORiN4cDAIDfpnL1nKabGc9pwq2KZ+LgUjynCeXRb/I5TQAAAOUVoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAACwhNAAAAFhCaAAAALCA0AQAAWEBoAgAAsIDQBAAAYAGhCQAAwAJCEwAAgAWEJgAAAAsITQAAABYQmgAAACwgNAEAAFhAaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWVPB2A/DUcszb3m4B5Uj6S3293QIA4P9wpQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABggVdD06ZNm/T73/9eoaGhstlsWrFihce4YRiaNGmSateuLX9/f8XGxurw4cMeNT/88IN69+4tp9OpqlWrauDAgTp37pxHze7du3XPPffIz89PYWFhmjFjRrFeli9frkaNGsnPz09RUVH6+OOPy3x/AQDAzcuroen8+fNq1qyZXn311RLHZ8yYoVdeeUULFizQ1q1bFRAQIJfLpQsXLpg1vXv31r59+5SSkqJVq1Zp06ZNGjJkiDnudrvVqVMnhYeHKz09XS+99JKmTJmiN954w6zZsmWLHnvsMQ0cOFCff/65unXrpm7dumnv3r2/3s4DAICbilf/jEqXLl3UpUuXEscMw9Ds2bM1ceJEPfzww5Kkt99+W8HBwVqxYoV69eqlAwcOKDk5Wdu3b1erVq0kSXPnztVDDz2k//mf/1FoaKgWL16svLw8/eMf/5DdbleTJk20c+dOvfzyy2a4mjNnjjp37qwxY8ZIkp577jmlpKRo3rx5WrBgwQ04EgAAoLwrt/c0HT16VJmZmYqNjTWXBQYGKjo6WmlpaZKktLQ0Va1a1QxMkhQbGysfHx9t3brVrLn33ntlt9vNGpfLpUOHDunMmTNmzaXbKaop2g4AAEC5/YO9mZmZkqTg4GCP5cHBweZYZmamgoKCPMYrVKig6tWre9TUq1ev2BxFY9WqVVNmZuYVt1OS3Nxc5ebmmu/dbve17B4AALjJlNsrTeXd9OnTFRgYaL7CwsK83RIAAPgVldvQFBISIknKysryWJ6VlWWOhYSE6OTJkx7jFy9e1A8//OBRU9Icl27jl2qKxksyYcIE5eTkmK9jx45d6y4CAICbSLkNTfXq1VNISIhSU1PNZW63W1u3blVMTIwkKSYmRtnZ2UpPTzdr1q1bp8LCQkVHR5s1mzZtUn5+vlmTkpKihg0bqlq1ambNpdspqinaTkkcDoecTqfHCwAA3Lq8GprOnTunnTt3aufOnZJ+vvl7586dysjIkM1m06hRo/T8889r5cqV2rNnj/r27avQ0FB169ZNktS4cWN17txZgwcP1rZt27R582YNHz5cvXr1UmhoqCTp8ccfl91u18CBA7Vv3z4tXbpUc+bMUUJCgtnHk08+qeTkZM2cOVMHDx7UlClTtGPHDg0fPvxGHxIAAFBOefVG8B07dqhDhw7m+6Ig069fPyUmJmrs2LE6f/68hgwZouzsbLVv317Jycny8/Mz11m8eLGGDx+uBx54QD4+PurevbteeeUVczwwMFBr165VfHy8WrZsqZo1a2rSpEkez3Jq27atlixZookTJ+qZZ55RgwYNtGLFCt1555034CgAAICbgc0wDMPbTdwK3G63AgMDlZOTc10f1bUc83YZdoWbXfpLfb3dAuckPHBOojy6nvPyWn5/l9t7mgAAAMoTQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAACwhNAAAAFhCaAAAALCA0AQAAWEBoAgAAsIDQBAAAYAGhCQAAwAJCEwAAgAWEJgAAAAsITQAAABYQmgAAACwgNAEAAFhAaAIAALCA0AQAAGABoQkAAMACQhMAAIAFhCYAAAALCE0AAAAWEJoAAAAsIDQBAABYQGgCAACwgNAEAABgAaEJAADAAkITAACABYQmAAAACwhNAAAAFhCaAAAALCA0AQAAWEBousyrr76qiIgI+fn5KTo6Wtu2bfN2SwAAoBwgNF1i6dKlSkhI0OTJk/XZZ5+pWbNmcrlcOnnypLdbAwAAXkZousTLL7+swYMHq3///oqMjNSCBQtUqVIl/eMf//B2awAAwMsITf8nLy9P6enpio2NNZf5+PgoNjZWaWlpXuwMAACUBxW83UB58f3336ugoEDBwcEey4ODg3Xw4MFi9bm5ucrNzTXf5+TkSJLcbvd19VGQ+9N1rY9by/WeT2WBcxKX4pxEeXQ952XRuoZhXLWW0FRK06dP17PPPltseVhYmBe6wa0qcO5Qb7cAeOCcRHlUFufl2bNnFRgYeMUaQtP/qVmzpnx9fZWVleWxPCsrSyEhIcXqJ0yYoISEBPN9YWGhfvjhB9WoUUM2m+1X7/dW5na7FRYWpmPHjsnpdHq7HYBzEuUO52TZMQxDZ8+eVWho6FVrCU3/x263q2XLlkpNTVW3bt0k/RyEUlNTNXz48GL1DodDDofDY1nVqlVvQKe/HU6nkx8GKFc4J1HecE6WjatdYSpCaLpEQkKC+vXrp1atWunuu+/W7Nmzdf78efXv39/brQEAAC8jNF2iZ8+eOnXqlCZNmqTMzEw1b95cycnJxW4OBwAAvz2EpssMHz68xI/jcOM4HA5Nnjy52MefgLdwTqK84Zz0Dpth5Tt2AAAAv3E83BIAAMACQhMAAIAFhCYAAAALCE246UVERGj27NnebgMolQ0bNshmsyk7O9vbreAmYPV84efir4PQhCt64oknZLPZ9OKLL3osX7FixQ1/8nliYmKJDxDdvn27hgwZckN7Qflzo87Vr7/+WjabTTt37iyzOXHrKTofbTab7Ha76tevr6lTp+rixYvXNW/btm114sQJ82GM/Fy8sQhNuCo/Pz/97W9/05kzZ7zdSolq1aqlSpUqebsNlAPl6VzNy8vzdgvwss6dO+vEiRM6fPiwnnrqKU2ZMkUvvfTSdc1pt9sVEhJy1f8I8HPx10FowlXFxsYqJCRE06dP/8WaTz/9VPfcc4/8/f0VFhamkSNH6vz58+b4iRMnFBcXJ39/f9WrV09Lliwpdvn45ZdfVlRUlAICAhQWFqa//OUvOnfunKSfL0n3799fOTk55v/epkyZIsnzMvTjjz+unj17evSWn5+vmjVr6u2335b085/HmT59uurVqyd/f381a9ZM7733XhkcKXhbWZyrNptNK1as8FinatWqSkxMlCTVq1dPknTXXXfJZrPp/vvvl/TzlYVu3bpp2rRpCg0NVcOGDSVJ//znP9WqVStVqVJFISEhevzxx3Xy5Mmy22mUWw6HQyEhIQoPD9ewYcMUGxurlStX6syZM+rbt6+qVaumSpUqqUuXLjp8+LC53jfffKPf//73qlatmgICAtSkSRN9/PHHkjw/nuPn4o1HaMJV+fr66oUXXtDcuXP17bffFhs/cuSIOnfurO7du2v37t1aunSpPv30U4+HhPbt21fHjx/Xhg0b9P777+uNN94o9ovDx8dHr7zyivbt26e33npL69at09ixYyX9fEl69uzZcjqdOnHihE6cOKGnn366WC+9e/fWRx99ZIYtSVqzZo1+/PFH/fGPf5QkTZ8+XW+//bYWLFigffv2afTo0frzn/+sjRs3lsnxgveUxbl6Ndu2bZMkffLJJzpx4oT+9a9/mWOpqak6dOiQUlJStGrVKkk//3J67rnntGvXLq1YsUJff/21nnjiievbUdyU/P39lZeXpyeeeEI7duzQypUrlZaWJsMw9NBDDyk/P1+SFB8fr9zcXG3atEl79uzR3/72N1WuXLnYfPxc9AIDuIJ+/foZDz/8sGEYhtGmTRtjwIABhmEYxgcffGAUnT4DBw40hgwZ4rHev//9b8PHx8f46aefjAMHDhiSjO3bt5vjhw8fNiQZs2bN+sVtL1++3KhRo4b5ftGiRUZgYGCxuvDwcHOe/Px8o2bNmsbbb79tjj/22GNGz549DcMwjAsXLhiVKlUytmzZ4jHHwIEDjccee+zKBwPlWlmcq4ZhGJKMDz74wKMmMDDQWLRokWEYhnH06FFDkvH5558X235wcLCRm5t7xT63b99uSDLOnj1rGIZhrF+/3pBknDlz5hr3GOXZpedjYWGhkZKSYjgcDqNbt26GJGPz5s1m7ffff2/4+/sby5YtMwzDMKKioowpU6aUOO/l5ws/F28s/owKLPvb3/6mjh07FvufzK5du7R7924tXrzYXGYYhgoLC3X06FF98cUXqlChglq0aGGO169fX9WqVfOY55NPPtH06dN18OBBud1uXbx4URcuXNCPP/5o+bP5ChUq6E9/+pMWL16sPn366Pz58/rwww/17rvvSpK+/PJL/fjjj3rwwQc91svLy9Ndd911TccD5Vdpz9XGjRtf13ajoqJkt9s9lqWnp2vKlCnatWuXzpw5o8LCQklSRkaGIiMjr2t7KN9WrVqlypUrKz8/X4WFhXr88cf1yCOPaNWqVYqOjjbratSooYYNG+rAgQOSpJEjR2rYsGFau3atYmNj1b17dzVt2rTUffBzsewQmmDZvffeK5fLpQkTJnh8vHDu3Dn913/9l0aOHFlsnbp16+qLL7646txff/21unbtqmHDhmnatGmqXr26Pv30Uw0cOFB5eXnXdENj7969dd999+nkyZNKSUmRv7+/OnfubPYqSUlJSbrttts81uNvON06SnuuSj/f02Rc9telij42uZqAgACP9+fPn5fL5ZLL5dLixYtVq1YtZWRkyOVycaP4b0CHDh302muvyW63KzQ0VBUqVNDKlSuvut6gQYPkcrmUlJSktWvXavr06Zo5c6ZGjBhR6l74uVg2CE24Ji+++KKaN29u3uQqSS1atND+/ftVv379Etdp2LChLl68qM8//1wtW7aU9PP/bC79hlN6eroKCws1c+ZM+fj8fKvdsmXLPOax2+0qKCi4ao9t27ZVWFiYli5dqtWrV+vRRx9VxYoVJUmRkZFyOBzKyMjQfffdd207j5tKac5V6edvHZ04ccJ8f/jwYf3444/m+6IrSVbOxYMHD+r06dN68cUXFRYWJknasWPHNe8Lbk4BAQHFzrXGjRvr4sWL2rp1q9q2bStJOn36tA4dOuRx5TEsLExDhw7V0KFDNWHCBL355pslhiZ+Lt5YhCZck6ioKPXu3VuvvPKKuWzcuHFq06aNhg8frkGDBikgIED79+9XSkqK5s2bp0aNGik2NlZDhgzRa6+9pooVK+qpp56Sv7+/+bXZ+vXrKz8/X3PnztXvf/97bd68WQsWLPDYdkREhM6dO6fU1FQ1a9ZMlSpV+sUrUI8//rgWLFigL774QuvXrzeXV6lSRU8//bRGjx6twsJCtW/fXjk5Odq8ebOcTqf69ev3Kxw1eENpzlVJ6tixo+bNm6eYmBgVFBRo3Lhx5i8XSQoKCpK/v7+Sk5NVp04d+fn5mc/MuVzdunVlt9s1d+5cDR06VHv37tVzzz336+44yrUGDRro4Ycf1uDBg/X666+rSpUqGj9+vG677TY9/PDDkqRRo0apS5cu+t3vfqczZ85o/fr1v/jRMT8XbzAv31OFcu7SmxmLHD161LDb7calp8+2bduMBx980KhcubIREBBgNG3a1Jg2bZo5fvz4caNLly6Gw+EwwsPDjSVLlhhBQUHGggULzJqXX37ZqF27tuHv72+4XC7j7bffLnaD7NChQ40aNWoYkozJkycbhuF5w2OR/fv3G5KM8PBwo7Cw0GOssLDQmD17ttGwYUOjYsWKRq1atQyXy2Vs3Ljx+g4WvKqsztXvvvvO6NSpkxEQEGA0aNDA+Pjjjz1uBDcMw3jzzTeNsLAww8fHx7jvvvt+cfuGYRhLliwxIiIiDIfDYcTExBgrV670uJGcG8FvTb90PhiGYfzwww9Gnz59jMDAQPPn3RdffGGODx8+3LjjjjsMh8Nh1KpVy+jTp4/x/fffG4ZR8vnCz8Ubx2YYl314D9wA3377rcLCwvTJJ5/ogQce8HY7AABcFaEJN8S6det07tw5RUVF6cSJExo7dqy+++47ffHFFx4ffQAAUF5xTxNuiPz8fD3zzDP66quvVKVKFbVt21aLFy8mMAEAbhpcaQIAALCAP6MCAABgAaEJAADAAkITAACABYQmAAAACwhNAFCCiIgIzZ4929ttAChHCE0AftMSExNVtWrVYsu3b9+uIUOG3PiGLrNhwwbZbDZlZ2d7uxXgN4/nNAFACWrVquXtFgCUM1xpAlDuvffee4qKipK/v79q1Kih2NhYnT9/XpK0cOFCNW7cWH5+fmrUqJHmz59vrvf111/LZrPpX//6lzp06KBKlSqpWbNmSktLk/TzVZz+/fsrJydHNptNNptNU6ZMkVT84zmbzabXX39dXbt2VaVKldS4cWOlpaXpyy+/1P3336+AgAC1bdtWR44c8ej9ww8/VIsWLeTn56fbb79dzz77rC5evOgx78KFC/XHP/5RlSpVUoMGDbRy5Uqz/w4dOkiSqlWrJpvNpieeeKKsDy8Aq7z5h+8A4GqOHz9uVKhQwXj55ZeNo0ePGrt37zZeffVV4+zZs8Y777xj1K5d23j//feNr776ynj//feN6tWrG4mJiYZh/PwHeyUZjRo1MlatWmUcOnTI6NGjhxEeHm7k5+cbubm5xuzZsw2n02mcOHHCOHHihHH27FnDMIr/wVNJxm233WYsXbrUOHTokNGtWzcjIiLC6Nixo5GcnGzs37/faNOmjdG5c2dznU2bNhlOp9NITEw0jhw5Yqxdu9aIiIgwpkyZ4jFvnTp1jCVLlhiHDx82Ro4caVSuXNk4ffq0cfHiReP99983JBmHDh0yTpw4YWRnZ9+YAw+gGEITgHItPT3dkGR8/fXXxcbuuOMOY8mSJR7LnnvuOSMmJsYwjP8fmhYuXGiO79u3z5BkHDhwwDAMw1i0aJERGBhYbO6SQtPEiRPN92lpaYYk4+9//7u57H//938NPz8/8/0DDzxgvPDCCx7z/vOf/zRq1679i/OeO3fOkGSsXr3aMIyS/6o9AO/gniYA5VqzZs30wAMPKCoqSi6XS506dVKPHj1kt9t15MgRDRw4UIMHDzbrL168qMDAQI85mjZtav67du3akqSTJ0+qUaNG19TLpfMEBwdLkqKiojyWXbhwQW63W06nU7t27dLmzZs1bdo0s6agoEAXLlzQjz/+qEqVKhWbNyAgQE6nUydPnrym3gD8+ghNAMo1X19fpaSkaMuWLVq7dq3mzp2rv/71r/roo48kSW+++aaio6OLrXOpS/8wtM1mkyQVFhZecy8lzXOluc+dO6dnn31WjzzySLG5/Pz8Spy3aJ7S9Afg10VoAlDu2Ww2tWvXTu3atdOkSZMUHh6uzZs3KzQ0VF999ZV69+5d6rntdrsKCgrKsNv/r0WLFjp06JDq169f6jnsdrsk/Wo9ArCO0ASgXNu6datSU1PVqVMnBQUFaevWrTp16pQaN26sZ599ViNHjlRgYKA6d+6s3Nxc7dixQ2fOnFFCQoKl+SMiInTu3DmlpqaqWbNmqlSpkvmx2fWaNGmSunbtqrp166pHjx7y8fHRrl27tHfvXj3//POW5ggPD5fNZtOqVav00EMPyd/fX5UrVy6T/gBcGx45AKBcczqd2rRpkx566CH97ne/08SJEzVz5kx16dJFgwYN0sKFC7Vo0SJFRUXpvvvuU2JiourVq2d5/rZt22ro0KHq2bOnatWqpRkzZpRZ7y6XS6tWrdLatWvVunVrtWnTRrNmzVJ4eLjlOW677TY9++yzGj9+vIKDgzV8+PAy6w/AtbEZhmF4uwkAAIDyjitNAAAAFhCaAAAALCA0AQAAWEBoAgAAsIDQBAAAYAGhCQAAwAJCEwAAgAWEJgAAAAsITQAAABYQmgAAACwgNAEAAFhAaAIAALDg/wGVOn2xaxgxCAAAAABJRU5ErkJggg==\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Max token length: 154\n",
            "95th percentile length: 18.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Build vocabulary\n",
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "# Join tokens for Keras Tokenizer\n",
        "texts = balanced_df['tokens'].apply(lambda x: ' '.join(x))\n",
        "\n",
        "tokenizer = Tokenizer()\n",
        "tokenizer.fit_on_texts(texts)\n",
        "sequences = tokenizer.texts_to_sequences(texts)\n",
        "word_index = tokenizer.word_index\n",
        "\n",
        "\n",
        "# Padding\n",
        "max_len = max(len(seq) for seq in sequences)\n",
        "X = pad_sequences(sequences, maxlen=max_len)\n",
        "y = pd.get_dummies(balanced_df['label']).values"
      ],
      "metadata": {
        "id": "rI9kFAM756R3"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# --- Load FastText Assamese Embeddings ---\n",
        "# Download from https://fasttext.cc/docs/en/crawl-vectors.html if not already present\n",
        "# Place the .vec or .bin file in your drive and set the path below\n",
        "!pip install --upgrade gensim\n",
        "from gensim.models import KeyedVectors\n",
        "import os\n",
        "\n",
        "fasttext_path = '/content/drive/MyDrive/cc.as.300.vec'  # Update this path to your FastText Assamese .vec file\n",
        "embedding_dim = 300  # FastText vectors are usually 300d\n",
        "\n",
        "if os.path.exists(fasttext_path):\n",
        "    ft_model = KeyedVectors.load_word2vec_format(fasttext_path)\n",
        "    print('FastText loaded')\n",
        "    vocab_size = len(word_index) + 1\n",
        "    embedding_matrix = np.zeros((vocab_size, embedding_dim))\n",
        "    for word, i in word_index.items():\n",
        "        if word in ft_model:\n",
        "            embedding_matrix[i] = ft_model[word]\n",
        "else:\n",
        "    print('FastText Assamese vectors not found, using random embeddings.')\n",
        "    embedding_matrix = None\n",
        "    embedding_dim = 128"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HfCD0tG3aBR4",
        "outputId": "f18767c3-e901-4ca0-f262-6a2849ff5750"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Requirement already satisfied: gensim in /usr/local/lib/python3.11/dist-packages (4.3.3)\n",
            "Requirement already satisfied: numpy<2.0,>=1.18.5 in /usr/local/lib/python3.11/dist-packages (from gensim) (1.26.4)\n",
            "Requirement already satisfied: scipy<1.14.0,>=1.7.0 in /usr/local/lib/python3.11/dist-packages (from gensim) (1.13.1)\n",
            "Requirement already satisfied: smart-open>=1.8.1 in /usr/local/lib/python3.11/dist-packages (from gensim) (7.3.0.post1)\n",
            "Requirement already satisfied: wrapt in /usr/local/lib/python3.11/dist-packages (from smart-open>=1.8.1->gensim) (1.17.2)\n",
            "FastText loaded\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.preprocessing.text import Tokenizer\n",
        "from tensorflow.keras.preprocessing.sequence import pad_sequences\n",
        "\n",
        "# Join tokens for Keras Tokenizer\n",
        "texts = balanced_df['tokens'].apply(lambda x: ' '.join(x))\n",
        "\n",
        "tokenizer = Tokenizer()\n",
        "tokenizer.fit_on_texts(texts)\n",
        "sequences = tokenizer.texts_to_sequences(texts)\n",
        "word_index = tokenizer.word_index\n",
        "\n",
        "max_len = max(len(seq) for seq in sequences)\n",
        "X = pad_sequences(sequences, maxlen=max_len)\n",
        "y = pd.get_dummies(balanced_df['label']).values"
      ],
      "metadata": {
        "id": "OXaHwWL1H9bw"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Train/test split\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)"
      ],
      "metadata": {
        "id": "toQdMWQy6AaA"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.layers import Layer, Input, Embedding, LSTM, Dense, Bidirectional, Dropout, BatchNormalization, SpatialDropout1D\n",
        "from tensorflow.keras.models import Model\n",
        "from tensorflow.keras import regularizers\n",
        "import tensorflow as tf\n",
        "\n",
        "# Attention Layer\n",
        "class Attention(Layer):\n",
        "    def __init__(self, **kwargs):\n",
        "        super(Attention, self).__init__(**kwargs)\n",
        "\n",
        "    def build(self, input_shape):\n",
        "        self.W = self.add_weight(name='att_weight', shape=(input_shape[-1], 1), initializer='normal')\n",
        "        self.b = self.add_weight(name='att_bias', shape=(input_shape[1], 1), initializer='zeros')\n",
        "        super(Attention, self).build(input_shape)\n",
        "\n",
        "    def call(self, x):\n",
        "        e = tf.keras.backend.tanh(tf.keras.backend.dot(x, self.W) + self.b)\n",
        "        a = tf.keras.backend.softmax(e, axis=1)\n",
        "        output = x * a\n",
        "        return tf.keras.backend.sum(output, axis=1)\n",
        "\n",
        "# Model definition\n",
        "input_ = Input(shape=(max_len,))\n",
        "\n",
        "if embedding_matrix is not None:\n",
        "    x = Embedding(vocab_size, embedding_dim, weights=[embedding_matrix], input_length=max_len, trainable=True)(input_)\n",
        "else:\n",
        "    x = Embedding(vocab_size, embedding_dim, input_length=max_len)(input_)\n",
        "\n",
        "x = SpatialDropout1D(0.3)(x)\n",
        "\n",
        "# Apply L2 to LSTM and Dense layers\n",
        "l2 = regularizers.l2(0.001)\n",
        "\n",
        "x = Bidirectional(LSTM(128, return_sequences=True, kernel_regularizer=l2, recurrent_regularizer=l2))(x)\n",
        "x = BatchNormalization()(x)\n",
        "x = Dropout(0.5)(x)\n",
        "\n",
        "x = Bidirectional(LSTM(64, return_sequences=True, kernel_regularizer=l2, recurrent_regularizer=l2))(x)\n",
        "x = Dropout(0.5)(x)\n",
        "\n",
        "x = Attention()(x)\n",
        "\n",
        "x = Dense(64, activation='relu', kernel_regularizer=l2)(x)\n",
        "output = Dense(y.shape[1], activation='softmax', kernel_regularizer=l2)(x)\n",
        "\n",
        "model = Model(inputs=input_, outputs=output)\n",
        "model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])\n",
        "model.summary()\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 579
        },
        "id": "HUEwlm2Puanh",
        "outputId": "ddfe6c3b-664a-4676-a299-7c5930cb4c42"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.11/dist-packages/keras/src/layers/core/embedding.py:90: UserWarning: Argument `input_length` is deprecated. Just remove it.\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1mModel: \"functional\"\u001b[0m\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional\"</span>\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
              "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                   \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape          \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m      Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
              "│ input_layer (\u001b[38;5;33mInputLayer\u001b[0m)        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m92\u001b[0m)             │             \u001b[38;5;34m0\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ embedding (\u001b[38;5;33mEmbedding\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m92\u001b[0m, \u001b[38;5;34m300\u001b[0m)        │     \u001b[38;5;34m9,570,000\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ spatial_dropout1d               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m92\u001b[0m, \u001b[38;5;34m300\u001b[0m)        │             \u001b[38;5;34m0\u001b[0m │\n",
              "│ (\u001b[38;5;33mSpatialDropout1D\u001b[0m)              │                        │               │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ bidirectional (\u001b[38;5;33mBidirectional\u001b[0m)   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m92\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │       \u001b[38;5;34m439,296\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ batch_normalization             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m92\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │         \u001b[38;5;34m1,024\u001b[0m │\n",
              "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)            │                        │               │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dropout (\u001b[38;5;33mDropout\u001b[0m)               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m92\u001b[0m, \u001b[38;5;34m256\u001b[0m)        │             \u001b[38;5;34m0\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ bidirectional_1 (\u001b[38;5;33mBidirectional\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m92\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │       \u001b[38;5;34m164,352\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m)             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m92\u001b[0m, \u001b[38;5;34m128\u001b[0m)        │             \u001b[38;5;34m0\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ attention (\u001b[38;5;33mAttention\u001b[0m)           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)            │           \u001b[38;5;34m220\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dense (\u001b[38;5;33mDense\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m)             │         \u001b[38;5;34m8,256\u001b[0m │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m)              │           \u001b[38;5;34m195\u001b[0m │\n",
              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
              "┃<span style=\"font-weight: bold\"> Layer (type)                    </span>┃<span style=\"font-weight: bold\"> Output Shape           </span>┃<span style=\"font-weight: bold\">       Param # </span>┃\n",
              "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
              "│ input_layer (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>)        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">92</span>)             │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ embedding (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Embedding</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">92</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">300</span>)        │     <span style=\"color: #00af00; text-decoration-color: #00af00\">9,570,000</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ spatial_dropout1d               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">92</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">300</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">SpatialDropout1D</span>)              │                        │               │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ bidirectional (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Bidirectional</span>)   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">92</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │       <span style=\"color: #00af00; text-decoration-color: #00af00\">439,296</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ batch_normalization             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">92</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │         <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
              "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)            │                        │               │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">92</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ bidirectional_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Bidirectional</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">92</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │       <span style=\"color: #00af00; text-decoration-color: #00af00\">164,352</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dropout_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">92</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ attention (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Attention</span>)           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)            │           <span style=\"color: #00af00; text-decoration-color: #00af00\">220</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)             │         <span style=\"color: #00af00; text-decoration-color: #00af00\">8,256</span> │\n",
              "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
              "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)              │           <span style=\"color: #00af00; text-decoration-color: #00af00\">195</span> │\n",
              "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m10,183,343\u001b[0m (38.85 MB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">10,183,343</span> (38.85 MB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m10,182,831\u001b[0m (38.84 MB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">10,182,831</span> (38.84 MB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m512\u001b[0m (2.00 KB)\n"
            ],
            "text/html": [
              "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> (2.00 KB)\n",
              "</pre>\n"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from tensorflow.keras.callbacks import ReduceLROnPlateau, EarlyStopping\n",
        "\n",
        "# --- Callbacks ---\n",
        "early_stop = EarlyStopping(\n",
        "    monitor='val_loss',\n",
        "    patience=4,\n",
        "    restore_best_weights=True,\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "reduce_lr = ReduceLROnPlateau(\n",
        "    monitor='val_loss',\n",
        "    factor=0.5,\n",
        "    patience=2,\n",
        "    min_lr=1e-6,\n",
        "    verbose=1\n",
        ")\n",
        "\n",
        "# --- Batch size tuned for dataset size and Colab TPU/GPU ---\n",
        "batch_size = 600  # Acceptable if memory allows; can drop to 128/256 for more stability\n",
        "\n",
        "# --- Model training ---\n",
        "history = model.fit(\n",
        "    X_train, y_train,\n",
        "    validation_data=(X_test, y_test),\n",
        "    epochs=10,\n",
        "    batch_size=batch_size,\n",
        "    callbacks=[early_stop, reduce_lr],\n",
        "    verbose=2\n",
        ")\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bvnmh0PKgdKi",
        "outputId": "b516f0ec-cdb2-49f0-a6dc-b45e426d7b5d"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Epoch 1/10\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import matplotlib.pyplot as plt\n",
        "\n",
        "acc = history.history['accuracy']\n",
        "val_acc = history.history['val_accuracy']\n",
        "loss = history.history['loss']\n",
        "val_loss = history.history['val_loss']\n",
        "\n",
        "epochs = range(1, len(acc) + 1)\n",
        "\n",
        "plt.plot(epochs, acc, 'b-', label='Training acc')\n",
        "plt.plot(epochs, val_acc, 'r', label='Validation acc')\n",
        "plt.title('Training and validation accuracy')\n",
        "plt.legend()\n",
        "\n",
        "plt.figure()\n",
        "\n",
        "plt.plot(epochs, loss, 'b-', label='Training loss')\n",
        "plt.plot(epochs, val_loss, 'r', label='Validation loss')\n",
        "plt.title('Training and validation loss')\n",
        "plt.legend()\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 887
        },
        "id": "oCjjjdfAfLIM",
        "outputId": "59e47322-473c-4ec5-d812-72df8a921679"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# --- Step 4: Evaluation ---\n",
        "from sklearn.metrics import classification_report, confusion_matrix\n",
        "import numpy as np\n",
        "\n",
        "preds = model.predict(X_test)\n",
        "y_pred = np.argmax(preds, axis=1)\n",
        "y_true = np.argmax(y_test, axis=1)\n",
        "\n",
        "# Get the unique sorted labels from the balanced dataframe\n",
        "labels = sorted(balanced_df['label'].unique())\n",
        "\n",
        "print(classification_report(y_true, y_pred, target_names=le.classes_, labels=labels))\n",
        "\n",
        "sns.heatmap(confusion_matrix(y_true, y_pred, labels=labels), annot=True, fmt='d', xticklabels=le.classes_, yticklabels=le.classes_)\n",
        "plt.xlabel('Predicted')\n",
        "plt.ylabel('Actual')\n",
        "plt.title('Confusion Matrix')\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 698
        },
        "id": "2q2MNFZd6ejn",
        "outputId": "f189c631-3edc-4a83-c6b3-00337b184237"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[1m242/242\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 9ms/step\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "    Negative       0.53      0.65      0.58      2582\n",
            "     Neutral       0.51      0.34      0.41      2600\n",
            "    Positive       0.47      0.52      0.50      2543\n",
            "\n",
            "    accuracy                           0.50      7725\n",
            "   macro avg       0.50      0.50      0.50      7725\n",
            "weighted avg       0.50      0.50      0.50      7725\n",
            "\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.11/dist-packages/sklearn/metrics/_classification.py:2687: UserWarning: labels size, 3, does not match size of target_names, 4\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 640x480 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "folder_path = '/content/drive/MyDrive/Corpus'\n",
        "os.makedirs(folder_path, exist_ok=True)\n",
        "\n",
        "model_path = os.path.join(folder_path, 'ASSAMESE_SENTIMENT_ANALYSIS.h5')\n",
        "model.save(model_path)\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gObdX8PXgO12",
        "outputId": "598a69e1-e7ba-43d4-a82b-2c444bdeaa77"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n"
          ]
        }
      ]
    }
  ]
}