{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "lczbBLnH1bPb",
      "metadata": {
        "id": "lczbBLnH1bPb"
      },
      "source": [
        "# Echo Chamber Linguistics\n",
        "### Can YouTube Comment Text Predict Political Channel Ideology?\n",
        "**CBS NLP Final Project 2026**\n",
        "\n",
        "---\n",
        "\n",
        "## Pipeline overview\n",
        "\n",
        "| # | Cell | Purpose |\n",
        "|---|---|---|\n",
        "| 1 | Environment setup | Detect Colab vs local, mount GPU, load `DEVICE` |\n",
        "| 2 | Load + clean data | Read raw CSV, dedup, length filter, language filter |\n",
        "| 3 | Exploratory analysis | Channel distribution, length stats, EDA plots |\n",
        "| — | *Preprocessing rationale* | Markdown: explains `text_clean` vs `text_clean_baseline` |\n",
        "| 4 | Text preprocessing | Build two cleaned-text columns (baseline + debiased) |\n",
        "| 5 | Label engineering | Binary `left=0 / right=1`, top-25 words per class |\n",
        "| 6 | Train / test split | Stratified 80/20, shared by all supervised models |\n",
        "| 7 | **Model 1** — Complement NB + BoW | Probabilistic baseline |\n",
        "| 8 | **Model 2** — LinearSVC + TF-IDF (word + char) | Strongest classical model |\n",
        "| 8.5 | Cross-channel generalisation | Channel-identity ablation (baseline vs debiased) |\n",
        "| 9 | **Model 3** — Word2Vec + LR | Static embeddings |\n",
        "| 10 | **Model 4** — DistilBERT fine-tuned | Pre-trained transformer |\n",
        "| 11 | **Model 5** — GPT-4o-mini zero-shot + few-shot | LLM without fine-tuning |\n",
        "| 12 | Model comparison | Aggregated results table + bar chart |\n",
        "| 13 | Error analysis | Disagreements between LR and BERT, hard examples |\n",
        "| 14 | LDA topic modelling | Unsupervised topic discovery per ideology |\n",
        "\n",
        "**Reproducibility.** All splits use `random_state=42`. Cells 7–13 share the same train/test split from Cell 6. Cell 8.5 builds its own cross-channel split. Cell 14 samples 30k per class.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "RnfWxA_dClOw",
      "metadata": {
        "id": "RnfWxA_dClOw"
      },
      "source": [
        "**Cell 1: Environment Setup**\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ErPo94Q51bPd",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ErPo94Q51bPd",
        "outputId": "9e5f99f5-849d-4f36-e8a8-d4ecad89e3ec"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Environment: Google Colab\n",
            "DATA_PATH = /root/.cache/huggingface/hub/datasets--Sapek007--yt-political-comments/snapshots/4eee51258acb7a95e99a8516e17ca569e9d28fe6/youtube_comments_clean.csv\n",
            "GPU: Tesla T4  |  VRAM: 15.6 GB  |  CUDA: 12.8\n",
            "PyTorch: 2.10.0+cu128  |  Device: cuda\n",
            "\n",
            "Package versions:\n",
            "  sklearn:       1.6.1\n",
            "  gensim:        4.4.0\n",
            "  transformers:  5.0.0\n",
            "  pandas:        2.2.2\n",
            "  numpy:         2.0.2\n"
          ]
        }
      ],
      "source": [
        "# Cell 1: Environment setup \n",
        "\n",
        "\n",
        "import sys, os, subprocess, warnings, logging\n",
        "\n",
        "\n",
        "subprocess.check_call([sys.executable, '-m', 'pip', 'install', '-q',\n",
        "                       'nltk', 'gensim', 'huggingface_hub'])\n",
        "\n",
        "\n",
        "warnings.filterwarnings('ignore', message='.*HF_TOKEN.*')\n",
        "logging.getLogger('huggingface_hub').setLevel(logging.ERROR)\n",
        "\n",
        "import torch\n",
        "from huggingface_hub import hf_hub_download\n",
        "\n",
        "IN_COLAB = 'google.colab' in sys.modules\n",
        "print(f'Environment: {\"Google Colab\" if IN_COLAB else \"Local machine\"}')\n",
        "\n",
        "\n",
        "DATA_PATH = hf_hub_download(\n",
        "    repo_id='Sapek007/yt-political-comments',\n",
        "    filename='youtube_comments_clean.csv',\n",
        "    repo_type='dataset',\n",
        ")\n",
        "print(f'DATA_PATH = {DATA_PATH}')\n",
        "\n",
        "if torch.cuda.is_available():\n",
        "    gpu_name = torch.cuda.get_device_name(0)\n",
        "    vram_gb  = torch.cuda.get_device_properties(0).total_memory / 1e9\n",
        "    print(f'GPU: {gpu_name}  |  VRAM: {vram_gb:.1f} GB  |  CUDA: {torch.version.cuda}')\n",
        "    DEVICE = torch.device('cuda')\n",
        "else:\n",
        "    print('No CUDA GPU detected — running on CPU.')\n",
        "    DEVICE = torch.device('cpu')\n",
        "\n",
        "print(f'PyTorch: {torch.__version__}  |  Device: {DEVICE}')\n",
        "\n",
        "\n",
        "try:\n",
        "    import sklearn, gensim, transformers, pandas, numpy\n",
        "    print(f'\\nPackage versions:')\n",
        "    print(f'  sklearn:       {sklearn.__version__}')\n",
        "    print(f'  gensim:        {gensim.__version__}')\n",
        "    print(f'  transformers:  {transformers.__version__}')\n",
        "    print(f'  pandas:        {pandas.__version__}')\n",
        "    print(f'  numpy:         {numpy.__version__}')\n",
        "except ImportError as e:\n",
        "    print(f'Some packages not yet installed: {e}')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "VC628damDjQb",
      "metadata": {
        "id": "VC628damDjQb"
      },
      "source": [
        "**Cell 2: Load + clean data**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "-y_ppnZm1bPe",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "-y_ppnZm1bPe",
        "outputId": "8fb27e93-ec76-4e07-86f9-221936c77dfe"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Raw rows: 375,435\n",
            "Removed 47,621 (text, channel) duplicates\n",
            "After strict English filter: 301,546\n",
            "\n",
            "Final rows: 301,546\n",
            "channel_name     channel_bias\n",
            "CNN              left            45533\n",
            "David Pakman     left            46559\n",
            "Fox News         right           44202\n",
            "PragerU          right           44593\n",
            "The Daily Wire   right           28062\n",
            "The Young Turks  left            45051\n",
            "Tim Pool         right           47546\n",
            "dtype: int64\n"
          ]
        }
      ],
      "source": [
        "# ── Cell 2: Load + clean data\n",
        "import pandas as pd\n",
        "\n",
        "df = pd.read_csv(DATA_PATH)\n",
        "print(f'Raw rows: {len(df):,}')\n",
        "\n",
        "df = df.drop(columns=['is_reply'])\n",
        "\n",
        "df.loc[df['channel_name'] == 'CNN', 'channel_bias'] = 'left'\n",
        "\n",
        "n_before = len(df)\n",
        "df = df.drop_duplicates(subset=['text',\n",
        "'channel_name']).reset_index(drop=True)\n",
        "print(f'Removed {n_before - len(df):,} (text, channel) duplicates')\n",
        "\n",
        "df = df[df['language'] == 'en'].reset_index(drop=True)\n",
        "print(f'After strict English filter: {len(df):,}')\n",
        "\n",
        "print(f'\\nFinal rows: {len(df):,}')\n",
        "print(df.groupby(['channel_name', 'channel_bias']).size())"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "Pp6zvGZbD0T0",
      "metadata": {
        "id": "Pp6zvGZbD0T0"
      },
      "source": [
        "**Cell 3: Exploratory Data Analysis**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e1dRVeUN1bPe",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "e1dRVeUN1bPe",
        "outputId": "e4647c99-7b10-484c-f3e8-256bd7bfb16a"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "=== Dataset after language filtering ===\n",
            "Raw rows (after dedup): 301,546\n",
            "English kept:           301,546\n",
            "Non-English dropped:    0\n",
            "\n",
            "=== Channel distribution ===\n",
            "   channel_name channel_bias  count\n",
            "            CNN         left  45533\n",
            "   David Pakman         left  46559\n",
            "       Fox News        right  44202\n",
            "        PragerU        right  44593\n",
            " The Daily Wire        right  28062\n",
            "The Young Turks         left  45051\n",
            "       Tim Pool        right  47546\n",
            "\n",
            "=== Bias distribution ===\n",
            "channel_bias\n",
            "right    164403\n",
            "left     137143\n",
            "Name: count, dtype: int64\n",
            "\n",
            "=== Comment length (words, raw text) ===\n",
            "                 count  mean   std  min  25%   50%   75%     max\n",
            "channel_bias                                                    \n",
            "left          137143.0  26.4  39.5  1.0  9.0  16.0  30.0  1804.0\n",
            "right         164403.0  25.9  38.6  1.0  9.0  16.0  29.0  1648.0\n",
            "\n",
            "1-2 word comments: 3,544 (1.2%)\n",
            "\n",
            "=== Missing values ===\n",
            "None\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1600x500 with 3 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Saved: eda_overview.png\n",
            "\n",
            "=== Like count by bias ===\n",
            "                 count  mean    std  min  25%  50%  75%      max\n",
            "channel_bias                                                    \n",
            "left          137143.0  8.00  89.32  0.0  0.0  0.0  2.0  11000.0\n",
            "right         164403.0  8.95  64.59  0.0  0.0  0.0  2.0   6000.0\n"
          ]
        }
      ],
      "source": [
        "# Cell 3: Exploratory Data Analysis\n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "from matplotlib.patches import Patch\n",
        "\n",
        "sns.set_theme(style='whitegrid', palette='muted', font_scale=1.1)\n",
        "COLORS = {'left': '#4C9BE8', 'right': '#E85C4C'}\n",
        "\n",
        "\n",
        "_df_en = df[df['language'] == 'en'].copy().reset_index(drop=True)\n",
        "_df_en['word_count'] = _df_en['text'].str.split().str.len()\n",
        "\n",
        "print('=== Dataset after language filtering ===')\n",
        "print(f'Raw rows (after dedup): {len(df):,}')\n",
        "print(f'English kept:           {len(_df_en):,}')\n",
        "print(f'Non-English dropped:    {len(df) - len(_df_en):,}')\n",
        "\n",
        "print(f'\\n=== Channel distribution ===')\n",
        "print(_df_en.groupby(['channel_name', 'channel_bias'])\n",
        "      .size().reset_index(name='count').to_string(index=False))\n",
        "\n",
        "print(f'\\n=== Bias distribution ===')\n",
        "print(_df_en['channel_bias'].value_counts())\n",
        "\n",
        "print(f'\\n=== Comment length (words, raw text) ===')\n",
        "print(_df_en.groupby('channel_bias')['word_count'].describe().round(1).to_string())\n",
        "print(f'\\n1-2 word comments: '\n",
        "      f'{(_df_en[\"word_count\"] <= 2).sum():,} '\n",
        "      f'({(_df_en[\"word_count\"] <= 2).mean()*100:.1f}%)')\n",
        "\n",
        "print(f'\\n=== Missing values ===')\n",
        "missing = _df_en.isnull().sum()\n",
        "print(missing[missing > 0].to_string() if missing.any() else 'None')\n",
        "\n",
        "fig, axes = plt.subplots(1, 3, figsize=(16, 5))\n",
        "fig.suptitle('Dataset overview (English comments, deduplicated)',\n",
        "             fontsize=14, fontweight='bold', y=1.02)\n",
        "\n",
        "\n",
        "ch = _df_en['channel_name'].value_counts()\n",
        "bar_colors = [COLORS[_df_en[_df_en['channel_name'] == c]['channel_bias'].iloc[0]]\n",
        "              for c in ch.index]\n",
        "axes[0].barh(ch.index, ch.values, color=bar_colors, edgecolor='white')\n",
        "axes[0].set_title('Comments per channel')\n",
        "axes[0].set_xlabel('Count')\n",
        "legend_els = [Patch(color=COLORS[b], label=b) for b in COLORS]\n",
        "axes[0].legend(handles=legend_els, loc='lower right', fontsize=9)\n",
        "\n",
        "\n",
        "bias_counts = _df_en['channel_bias'].value_counts()\n",
        "axes[1].pie(bias_counts.values,\n",
        "            labels=bias_counts.index,\n",
        "            colors=[COLORS[b] for b in bias_counts.index],\n",
        "            autopct='%1.1f%%', startangle=90,\n",
        "            wedgeprops={'edgecolor': 'white', 'linewidth': 2})\n",
        "axes[1].set_title('Bias distribution')\n",
        "\n",
        "\n",
        "for bias, grp in _df_en.groupby('channel_bias'):\n",
        "    grp['word_count'].clip(upper=80).plot.hist(\n",
        "        ax=axes[2], bins=40, alpha=0.5, color=COLORS[bias], label=bias)\n",
        "axes[2].set_title('Comment length by bias')\n",
        "axes[2].set_xlabel('Word count (capped at 80)')\n",
        "axes[2].legend()\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.savefig('eda_overview.png', dpi=150, bbox_inches='tight')\n",
        "plt.show()\n",
        "print('Saved: eda_overview.png')\n",
        "\n",
        "print('\\n=== Like count by bias ===')\n",
        "print(_df_en.groupby('channel_bias')['like_count'].describe().round(2).to_string())"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "PDNELiQAq7yi",
      "metadata": {
        "id": "PDNELiQAq7yi"
      },
      "source": [
        "## Preprocessing rationale\n",
        "\n",
        "Two cleaning passes are produced from the same raw text:\n",
        "\n",
        "- **`text_clean_baseline`** — removes only generic high-frequency stop words (`people`, `just`, `like`, `know`, ...). Politically charged terms (e.g. `trump`, `biden`) are intentionally retained — their usage pattern differs across left vs right and carries genuine ideological signal.\n",
        "- **`text_clean`** — additionally removes **channel identity tokens** (host names, brand names: `cenk`, `pakman`, `prageru`, `shapiro`, `tucker`, ...). Used in the cross-channel generalisation test to measure how much of the model's accuracy comes from ideology vs. surface channel branding.\n",
        "\n",
        "The contrast between these two conditions is the core of the cross-channel ablation in Cell 8.5.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "rWdZ1at4EDE6",
      "metadata": {
        "id": "rWdZ1at4EDE6"
      },
      "source": [
        "**Cell 4: Text Preprocessing**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "k77Z8fto1bPf",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "k77Z8fto1bPf",
        "outputId": "af6dd3fe-e780-4594-9b90-5e143bb7e5ec"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "English comments: 301,536\n",
            "Columns: ['text', 'text_clean_baseline', 'text_clean']\n",
            "\n",
            "Bias distribution:\n",
            "channel_bias\n",
            "right    164400\n",
            "left     137136\n",
            "Name: count, dtype: int64\n"
          ]
        }
      ],
      "source": [
        "#  Cell 4: Text Preprocessing\n",
        "\n",
        "import re\n",
        "\n",
        "# CUSTOM_STOP: high-frequency generic words with no ideological signal.\n",
        "CUSTOM_STOP = {\n",
        "    'people', 'just', 'like', 'don', 'know', 'think',\n",
        "    'going', 'good', 'want', 'did', 'say', 'does', 'time', 'need'\n",
        "}\n",
        "\n",
        "# CHANNEL_STOP: host names and channel brand tokens.\n",
        "CHANNEL_STOP = {\n",
        "    'ana', 'cenk', 'tyt', 'kasparian', 'uygur', 'anna', 'kent',\n",
        "    'tipping', 'tip', 'caller',\n",
        "    'david', 'pakman', 'pakmanshow',\n",
        "    'cnn', 'fareed', 'zakaria', 'anderson', 'cooper', 'enten',\n",
        "    'harry', 'bolton', 'hogg', 'abdul', 'powell',\n",
        "    'dennis', 'prager', 'prageru',\n",
        "    'timcast', 'tim',\n",
        "    'shapiro', 'candace', 'owens', 'walsh', 'knowles',\n",
        "    'malice', 'loomer', 'hegseth', 'pete', 'ian',\n",
        "    'charlie', 'kirk', 'crowder',\n",
        "    'fox', 'tucker', 'carlson', 'hannity', 'ingraham',\n",
        "}\n",
        "\n",
        "\n",
        "def clean_text(text, extra_stop=None):\n",
        "    \"\"\"Lowercase, strip URLs/punctuation, drop stop tokens.\"\"\"\n",
        "    if not isinstance(text, str):\n",
        "        return ''\n",
        "    text = text.lower()\n",
        "    text = re.sub(r'http\\S+|www\\S+', ' ', text)\n",
        "    text = re.sub(r'[^\\w\\s]', ' ', text)\n",
        "    text = re.sub(r'\\s+', ' ', text).strip()\n",
        "    stop = CUSTOM_STOP | (extra_stop or set())\n",
        "    tokens = [w for w in text.split() if w not in stop]\n",
        "    return ' '.join(tokens)\n",
        "\n",
        "\n",
        "\n",
        "df_en = df.copy().reset_index(drop=True)\n",
        "\n",
        "\n",
        "df_en['text_clean_baseline'] = df_en['text'].apply(\n",
        "    lambda t: clean_text(t, extra_stop=None))\n",
        "\n",
        "\n",
        "df_en['text_clean'] = df_en['text'].apply(\n",
        "    lambda t: clean_text(t, extra_stop=CHANNEL_STOP))\n",
        "\n",
        "\n",
        "df_en = df_en[df_en['text_clean'].str.len() > 0].reset_index(drop=True)\n",
        "\n",
        "print(f'English comments: {len(df_en):,}')\n",
        "print(f'Columns: {[c for c in df_en.columns if \"text\" in c]}')\n",
        "print(f'\\nBias distribution:')\n",
        "print(df_en['channel_bias'].value_counts())"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "2R4KmMhTEK49",
      "metadata": {
        "id": "2R4KmMhTEK49"
      },
      "source": [
        "**Cell 5: Label Engineering — 2-class (left / right)**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "toCc55RN1bPf",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 650
        },
        "id": "toCc55RN1bPf",
        "outputId": "cb42fa2e-d129-4cb6-9237-604c1c0561d6"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "2-class label distribution:\n",
            "  left  (TYT + Pakman + CNN): 137,136  (45.5%)\n",
            "  right (Fox + PragerU + Daily Wire + Tim Pool): 164,400  (54.5%)\n",
            "  Total: 301,536\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1400x700 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Saved: top_words_per_ideology.png\n"
          ]
        }
      ],
      "source": [
        "# Cell 5: Label Engineering — 2-class (left / right)\n",
        "\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "import numpy as np\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "LABEL_MAP    = {'left': 0, 'right': 1}\n",
        "TARGET_NAMES = ['left', 'right']\n",
        "\n",
        "df_model = df_en[df_en['channel_bias'].isin(LABEL_MAP)].copy()\n",
        "df_model['label'] = df_model['channel_bias'].map(LABEL_MAP)\n",
        "\n",
        "print(\"2-class label distribution:\")\n",
        "for name, v in [('left  (TYT + Pakman + CNN)', 0),\n",
        "                ('right (Fox + PragerU + Daily Wire + Tim Pool)', 1)]:\n",
        "    n   = (df_model['label'] == v).sum()\n",
        "    pct = n / len(df_model) * 100\n",
        "    print(f\"  {name}: {n:,}  ({pct:.1f}%)\")\n",
        "print(f\"  Total: {len(df_model):,}\")\n",
        "\n",
        "\n",
        "def top_words(texts, n=25):\n",
        "    vec = TfidfVectorizer(max_features=5000, stop_words='english', ngram_range=(1, 1))\n",
        "    X = vec.fit_transform(texts)\n",
        "    mean_tfidf = np.asarray(X.mean(axis=0)).flatten()\n",
        "    idx = mean_tfidf.argsort()[-n:][::-1]\n",
        "    return [vec.get_feature_names_out()[i] for i in idx], mean_tfidf[idx]\n",
        "\n",
        "fig, axes = plt.subplots(1, 2, figsize=(14, 7))\n",
        "palette = {'left': '#4C9BE8', 'right': '#E85C4C'}\n",
        "\n",
        "for ax, (bias, label_idx) in zip(axes, [('left', 0), ('right', 1)]):\n",
        "    texts = df_model[df_model['label'] == label_idx]['text_clean']\n",
        "    words, scores_w = top_words(texts)\n",
        "    ax.barh(words[::-1], scores_w[::-1], color=palette[bias], edgecolor='white')\n",
        "    ax.set_title(f'Top 25 words — {bias.upper()}', fontweight='bold')\n",
        "    ax.set_xlabel('Mean TF-IDF')\n",
        "\n",
        "plt.suptitle('Vocabulary Fingerprint by Ideology (left vs right)',\n",
        "             fontsize=14, fontweight='bold', y=1.01)\n",
        "plt.tight_layout()\n",
        "plt.savefig('top_words_per_ideology.png', dpi=150, bbox_inches='tight')\n",
        "plt.show()\n",
        "print('Saved: top_words_per_ideology.png')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "Mtg7YbkAET2W",
      "metadata": {
        "id": "Mtg7YbkAET2W"
      },
      "source": [
        "**Cell 6: Train / Test Split**\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "x0UuBqJa1bPg",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "x0UuBqJa1bPg",
        "outputId": "277aa914-0941-406a-e1c4-d4a81e66b853"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Train: 241,228  |  Test: 60,308\n",
            "Label distribution (train):\n",
            "  left: 109,708\n",
            "  right: 131,520\n"
          ]
        }
      ],
      "source": [
        "# Cell 6: Train / Test Split\n",
        "from sklearn.model_selection import train_test_split\n",
        "\n",
        "X = df_model['text_clean']\n",
        "y = df_model['label']\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(\n",
        "    X, y, test_size=0.2, random_state=42, stratify=y\n",
        ")\n",
        "\n",
        "print(f\"Train: {len(X_train):,}  |  Test: {len(X_test):,}\")\n",
        "print(\"Label distribution (train):\")\n",
        "for i, name in enumerate(TARGET_NAMES):\n",
        "    print(f\"  {name}: {(y_train == i).sum():,}\")"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "2LJGICds1bPg",
      "metadata": {
        "id": "2LJGICds1bPg"
      },
      "source": [
        "---\n",
        "## Models\n",
        "\n",
        "Five supervised/prompted classifiers are trained or evaluated on the same\n",
        "80/20 stratified split (Cell 6). Cell 8.5 sits between the linear model\n",
        "and the embedding model because it re-uses the LinearSVC pipeline for\n",
        "the cross-channel ablation.\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "85ZNdqkrEdam",
      "metadata": {
        "id": "85ZNdqkrEdam"
      },
      "source": [
        "**Cell 7: Model - Naive Bayes + Bag of Words (optimised)**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "7TV2qnQe1bPg",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 671
        },
        "id": "7TV2qnQe1bPg",
        "outputId": "16a11748-a7bd-4f52-b212-84ec5c78ff0f"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "=== Model 1: Naive Bayes + BoW (optimised) ===\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        left       0.66      0.72      0.69     27428\n",
            "       right       0.75      0.70      0.72     32880\n",
            "\n",
            "    accuracy                           0.71     60308\n",
            "   macro avg       0.71      0.71      0.71     60308\n",
            "weighted avg       0.71      0.71      0.71     60308\n",
            "\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 600x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Cell 7: Model 1 — Naive Bayes + Bag of Words (optimised)\n",
        "\n",
        "\n",
        "from sklearn.feature_extraction.text import CountVectorizer\n",
        "from sklearn.naive_bayes import ComplementNB\n",
        "from sklearn.metrics import classification_report, f1_score, accuracy_score, confusion_matrix, ConfusionMatrixDisplay\n",
        "\n",
        "bow = CountVectorizer(max_features=50000, ngram_range=(1, 2), min_df=2)\n",
        "X_train_bow = bow.fit_transform(X_train)\n",
        "X_test_bow  = bow.transform(X_test)\n",
        "\n",
        "nb = ComplementNB(alpha=0.1)\n",
        "nb.fit(X_train_bow, y_train)\n",
        "y_pred_nb = nb.predict(X_test_bow)\n",
        "\n",
        "acc_nb = accuracy_score(y_test, y_pred_nb)\n",
        "f1_nb  = f1_score(y_test, y_pred_nb, average='macro')\n",
        "\n",
        "print('=== Model 1: Naive Bayes + BoW (optimised) ===')\n",
        "print(classification_report(y_test, y_pred_nb, target_names=TARGET_NAMES))\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(6, 5))\n",
        "ConfusionMatrixDisplay(confusion_matrix(y_test, y_pred_nb), display_labels=TARGET_NAMES).plot(\n",
        "    ax=ax, colorbar=False, cmap='Blues')\n",
        "ax.set_title(f'Naive Bayes + BoW (optimised)\\nAcc={acc_nb:.3f}  F1={f1_nb:.3f}', fontweight='bold')\n",
        "plt.tight_layout(); plt.savefig('cm_naive_bayes.png', dpi=150, bbox_inches='tight'); plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "VrPuITzkEtWH",
      "metadata": {
        "id": "VrPuITzkEtWH"
      },
      "source": [
        "**Cell 8: Model 2 — LinearSVC + TF-IDF word + char n-grams (optimised)**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "ZcSZ000_1bPg",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "ZcSZ000_1bPg",
        "outputId": "fd1b548f-2c75-4107-c4ed-9ba29d7c0837"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "=== Model 2: LinearSVC + TF-IDF word + char n-grams (optimised) ===\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        left       0.68      0.71      0.70     27428\n",
            "       right       0.75      0.72      0.74     32880\n",
            "\n",
            "    accuracy                           0.72     60308\n",
            "   macro avg       0.72      0.72      0.72     60308\n",
            "weighted avg       0.72      0.72      0.72     60308\n",
            "\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 600x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1400x600 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "#Cell 8: Model 2 — LinearSVC + TF-IDF word + char n-grams (optimised)\n",
        "\n",
        "\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.svm import LinearSVC\n",
        "from sklearn.metrics import classification_report, f1_score, accuracy_score, confusion_matrix, ConfusionMatrixDisplay\n",
        "from scipy.sparse import hstack\n",
        "import numpy as np\n",
        "\n",
        "word_tfidf = TfidfVectorizer(max_features=80000, ngram_range=(1, 3),\n",
        "                              min_df=3, sublinear_tf=True, analyzer='word')\n",
        "char_tfidf = TfidfVectorizer(max_features=40000, ngram_range=(2, 4),\n",
        "                              min_df=5, sublinear_tf=True, analyzer='char_wb')\n",
        "\n",
        "X_train_w = word_tfidf.fit_transform(X_train)\n",
        "X_test_w  = word_tfidf.transform(X_test)\n",
        "X_train_c = char_tfidf.fit_transform(X_train)\n",
        "X_test_c  = char_tfidf.transform(X_test)\n",
        "\n",
        "X_train_combined = hstack([X_train_w, X_train_c])\n",
        "X_test_combined  = hstack([X_test_w,  X_test_c])\n",
        "\n",
        "svc = LinearSVC(class_weight='balanced', C=0.5, max_iter=2000)\n",
        "svc.fit(X_train_combined, y_train)\n",
        "y_pred_lr = svc.predict(X_test_combined)\n",
        "\n",
        "acc_lr = accuracy_score(y_test, y_pred_lr)\n",
        "f1_lr  = f1_score(y_test, y_pred_lr, average='macro')\n",
        "\n",
        "print('=== Model 2: LinearSVC + TF-IDF word + char n-grams (optimised) ===')\n",
        "print(classification_report(y_test, y_pred_lr, target_names=TARGET_NAMES))\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(6, 5))\n",
        "ConfusionMatrixDisplay(confusion_matrix(y_test, y_pred_lr), display_labels=TARGET_NAMES).plot(\n",
        "    ax=ax, colorbar=False, cmap='Oranges')\n",
        "ax.set_title(f'LinearSVC + TF-IDF word+char\\nAcc={acc_lr:.3f}  F1={f1_lr:.3f}', fontweight='bold')\n",
        "plt.tight_layout()\n",
        "plt.savefig('cm_lr_tfidf.png', dpi=150, bbox_inches='tight')\n",
        "plt.show()\n",
        "\n",
        "\n",
        "feature_names = np.concatenate([word_tfidf.get_feature_names_out(),\n",
        "                                 char_tfidf.get_feature_names_out()])\n",
        "coefs = svc.coef_[0]\n",
        "\n",
        "fig, axes = plt.subplots(1, 2, figsize=(14, 6))\n",
        "fig.suptitle('LinearSVC — Most Predictive Features per Class',\n",
        "             fontsize=14, fontweight='bold')\n",
        "\n",
        "\n",
        "top_left = np.argsort(coefs)[:15]\n",
        "axes[0].barh(feature_names[top_left], coefs[top_left],\n",
        "             color='#4C9BE8', edgecolor='white')\n",
        "axes[0].set_title('Top 15 → LEFT')\n",
        "axes[0].set_xlabel('Coefficient')\n",
        "\n",
        "\n",
        "top_right = np.argsort(coefs)[-15:][::-1]\n",
        "axes[1].barh(feature_names[top_right], coefs[top_right],\n",
        "             color='#E85C4C', edgecolor='white')\n",
        "axes[1].set_title('Top 15 → RIGHT')\n",
        "axes[1].set_xlabel('Coefficient')\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.savefig('lr_top_features.png', dpi=150, bbox_inches='tight')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "mB0r75jOExDW",
      "metadata": {
        "id": "mB0r75jOExDW"
      },
      "source": [
        "**Cell 8.5: Cross-channel generalisation test**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "Dfh0L87T1bPh",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "Dfh0L87T1bPh",
        "outputId": "297f67a6-64fa-44fa-b951-420564acd983"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Cross-channel train: 170,263\n",
            "label\n",
            "right    88792\n",
            "left     81471\n",
            "Name: count, dtype: int64\n",
            "\n",
            "Cross-channel test:  131,273\n",
            "label\n",
            "right    75608\n",
            "left     55665\n",
            "Name: count, dtype: int64\n",
            "\n",
            "=== Cross-channel — BASELINE (channel names present) ===\n",
            "Acc=0.593  F1=0.592\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        left       0.52      0.63      0.57     55665\n",
            "       right       0.67      0.57      0.62     75608\n",
            "\n",
            "    accuracy                           0.59    131273\n",
            "   macro avg       0.60      0.60      0.59    131273\n",
            "weighted avg       0.61      0.59      0.60    131273\n",
            "\n",
            "\n",
            "=== Cross-channel — DEBIASED (channel names removed) ===\n",
            "Acc=0.586  F1=0.585\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        left       0.51      0.64      0.57     55665\n",
            "       right       0.67      0.55      0.60     75608\n",
            "\n",
            "    accuracy                           0.59    131273\n",
            "   macro avg       0.59      0.59      0.58    131273\n",
            "weighted avg       0.60      0.59      0.59    131273\n",
            "\n",
            "\n",
            "==========================================================\n",
            "Condition                             F1 macro\n",
            "----------------------------------------------------------\n",
            "Within-channel (standard split)          0.718\n",
            "Cross-channel — baseline                 0.592\n",
            "Cross-channel — debiased                 0.585\n",
            "==========================================================\n",
            "Channel identity effect:  0.126\n",
            "Debiasing recovery:      +-0.007\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Saved: cross_channel_generalisation.png\n"
          ]
        }
      ],
      "source": [
        "#Cell 8.5: Cross-channel generalisation test\n",
        "\n",
        "\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.svm import LinearSVC\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import (classification_report, f1_score,\n",
        "                              accuracy_score, confusion_matrix,\n",
        "                              ConfusionMatrixDisplay)\n",
        "from scipy.sparse import hstack\n",
        "import matplotlib.pyplot as plt\n",
        "\n",
        "LABEL_MAP    = {'left': 0, 'right': 1}\n",
        "TARGET_NAMES = ['left', 'right']\n",
        "\n",
        "if 'text_clean_baseline' not in df_en.columns:\n",
        "    df_en['text_clean_baseline'] = df_en['text_clean']\n",
        "    print('WARNING: text_clean_baseline not found — using text_clean for both conditions.')\n",
        "\n",
        "\n",
        "train_parts, test_parts = [], []\n",
        "\n",
        "\n",
        "tyt = df_en[df_en['channel_name'] == 'The Young Turks'][\n",
        "    ['text_clean', 'text_clean_baseline']].copy()\n",
        "tyt['label'] = 0\n",
        "train_parts.append(tyt)\n",
        "\n",
        "\n",
        "for ch in ['Fox News', 'PragerU']:\n",
        "    sub = df_en[df_en['channel_name'] == ch][\n",
        "        ['text_clean', 'text_clean_baseline']].copy()\n",
        "    sub['label'] = 1\n",
        "    train_parts.append(sub)\n",
        "\n",
        "\n",
        "pakman = df_en[df_en['channel_name'] == 'David Pakman'][\n",
        "    ['text_clean', 'text_clean_baseline']].copy()\n",
        "pakman['label'] = 0\n",
        "test_parts.append(pakman)\n",
        "\n",
        "\n",
        "for ch in ['The Daily Wire', 'Tim Pool']:\n",
        "    sub = df_en[df_en['channel_name'] == ch][\n",
        "        ['text_clean', 'text_clean_baseline']].copy()\n",
        "    sub['label'] = 1\n",
        "    test_parts.append(sub)\n",
        "\n",
        "\n",
        "cnn_data = df_en[df_en['channel_name'] == 'CNN'][\n",
        "    ['text_clean', 'text_clean_baseline']].copy()\n",
        "cnn_data['label'] = 0\n",
        "cnn_train, cnn_test = train_test_split(cnn_data, test_size=0.2, random_state=42)\n",
        "train_parts.append(cnn_train)\n",
        "test_parts.append(cnn_test)\n",
        "\n",
        "train_cc = pd.concat(train_parts).sample(frac=1, random_state=42).reset_index(drop=True)\n",
        "test_cc  = pd.concat(test_parts).sample(frac=1, random_state=42).reset_index(drop=True)\n",
        "\n",
        "print(f'Cross-channel train: {len(train_cc):,}')\n",
        "print(train_cc['label'].value_counts().rename({0: 'left', 1: 'right'}))\n",
        "print(f'\\nCross-channel test:  {len(test_cc):,}')\n",
        "print(test_cc['label'].value_counts().rename({0: 'left', 1: 'right'}))\n",
        "\n",
        "\n",
        "CHANNEL_NAMES = {\n",
        "    'ana','cenk','tyt','kasparian','uygur','anna','kent','tipping','tip',\n",
        "    'caller','david','pakman','cnn','fareed','zakaria','anderson','cooper',\n",
        "    'enten','harry','bolton','hogg','powell','dennis','prager','prageru',\n",
        "    'timcast','tim','shapiro','candace','owens','walsh','malice','loomer',\n",
        "    'hegseth','pete','ian','charlie','kirk','crowder','fox','tucker',\n",
        "    'carlson','hannity','ingraham',\n",
        "}\n",
        "\n",
        "def run_svc(X_train_texts, X_test_texts, y_train, y_test,\n",
        "            label='', extra_stop=None):\n",
        "    sw = list(extra_stop) if extra_stop else None\n",
        "    wv = TfidfVectorizer(max_features=60000, ngram_range=(1, 3),\n",
        "                         min_df=3, sublinear_tf=True, analyzer='word',\n",
        "                         stop_words=sw)\n",
        "    cv = TfidfVectorizer(max_features=30000, ngram_range=(2, 4),\n",
        "                         min_df=5, sublinear_tf=True, analyzer='char_wb')\n",
        "    X_tr = hstack([wv.fit_transform(X_train_texts),\n",
        "                   cv.fit_transform(X_train_texts)])\n",
        "    X_te = hstack([wv.transform(X_test_texts),\n",
        "                   cv.transform(X_test_texts)])\n",
        "    clf  = LinearSVC(class_weight='balanced', C=0.5, max_iter=2000)\n",
        "    clf.fit(X_tr, y_train)\n",
        "    preds = clf.predict(X_te)\n",
        "    f1    = f1_score(y_test, preds, average='macro')\n",
        "    acc   = accuracy_score(y_test, preds)\n",
        "    print(f'\\n=== {label} ===')\n",
        "    print(f'Acc={acc:.3f}  F1={f1:.3f}')\n",
        "    print(classification_report(y_test, preds, target_names=TARGET_NAMES))\n",
        "    return preds, f1\n",
        "\n",
        "\n",
        "preds_baseline, f1_cross_baseline = run_svc(\n",
        "    train_cc['text_clean_baseline'],\n",
        "    test_cc['text_clean_baseline'],\n",
        "    train_cc['label'],\n",
        "    test_cc['label'],\n",
        "    label='Cross-channel — BASELINE (channel names present)'\n",
        ")\n",
        "\n",
        "\n",
        "preds_debiased, f1_cross_debiased = run_svc(\n",
        "    train_cc['text_clean'],\n",
        "    test_cc['text_clean'],\n",
        "    train_cc['label'],\n",
        "    test_cc['label'],\n",
        "    label='Cross-channel — DEBIASED (channel names removed)',\n",
        "    extra_stop=CHANNEL_NAMES\n",
        ")\n",
        "\n",
        "\n",
        "print('\\n' + '=' * 58)\n",
        "print(f'{\"Condition\":<35} {\"F1 macro\":>10}')\n",
        "print('-' * 58)\n",
        "print(f'{\"Within-channel (standard split)\":<35} {f1_lr:>10.3f}')\n",
        "print(f'{\"Cross-channel — baseline\":<35} {f1_cross_baseline:>10.3f}')\n",
        "print(f'{\"Cross-channel — debiased\":<35} {f1_cross_debiased:>10.3f}')\n",
        "print('=' * 58)\n",
        "print(f'Channel identity effect:  {f1_lr - f1_cross_baseline:.3f}')\n",
        "print(f'Debiasing recovery:      +{f1_cross_debiased - f1_cross_baseline:.3f}')\n",
        "\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(8, 5))\n",
        "labels = ['Within-channel\\n(standard)', 'Cross-channel\\n(baseline)',\n",
        "          'Cross-channel\\n(debiased)']\n",
        "values = [f1_lr, f1_cross_baseline, f1_cross_debiased]\n",
        "colors = ['#4C9BE8', '#E85C4C', '#7BC8A4']\n",
        "bars   = ax.bar(labels, values, color=colors, edgecolor='white', width=0.5)\n",
        "ax.set_ylim(0, 1.0)\n",
        "ax.set_ylabel('F1 macro')\n",
        "ax.set_title('Channel identity effect — LinearSVC generalisation (2-class)',\n",
        "             fontsize=13, fontweight='bold')\n",
        "ax.axhline(0.5, color='grey', linestyle='--', linewidth=1,\n",
        "           label='Random baseline (2-class)')\n",
        "ax.legend(fontsize=9)\n",
        "ax.grid(axis='y', alpha=0.4)\n",
        "for bar in bars:\n",
        "    ax.text(bar.get_x() + bar.get_width() / 2, bar.get_height() + 0.01,\n",
        "            f'{bar.get_height():.3f}', ha='center', fontsize=11, fontweight='bold')\n",
        "plt.tight_layout()\n",
        "plt.savefig('cross_channel_generalisation.png', dpi=150, bbox_inches='tight')\n",
        "plt.show()\n",
        "print('Saved: cross_channel_generalisation.png')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "lfcXiFDVE3RB",
      "metadata": {
        "id": "lfcXiFDVE3RB"
      },
      "source": [
        "**Cell 9: Model 3 — Word2Vec embeddings + Logistic Regression (optimised)**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "2KmZlXZd1bPh",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 899
        },
        "id": "2KmZlXZd1bPh",
        "outputId": "be544198-666a-44b2-edd4-87e00563a1bd"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m27.9/27.9 MB\u001b[0m \u001b[31m42.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hTraining Word2Vec (skip-gram)...\n",
            "Vocabulary: 31,031 words\n",
            "Vectorising train set...\n",
            "Vectorising test set...\n",
            "\n",
            "=== Model 3: Word2Vec skip-gram + TF-IDF weighted LR (optimised) ===\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        left       0.62      0.67      0.65     27428\n",
            "       right       0.71      0.66      0.68     32880\n",
            "\n",
            "    accuracy                           0.66     60308\n",
            "   macro avg       0.66      0.67      0.66     60308\n",
            "weighted avg       0.67      0.66      0.67     60308\n",
            "\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 600x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "=== Nearest to \"trump\" ===\n",
            "[('drowsy', 0.626954972743988), ('pedifile', 0.617436945438385), ('he', 0.5996278524398804), ('krasnov', 0.5820353031158447), ('junkie', 0.5723727941513062), ('hus', 0.5656329989433289), ('baffoon', 0.5652154684066772), ('bestest', 0.5620992183685303)]\n",
            "\n",
            "=== Nearest to \"biden\" ===\n",
            "[('bidens', 0.6565695405006409), ('joe', 0.6180587410926819), ('borelli', 0.6006743311882019), ('obama', 0.5693072080612183), ('autopen', 0.5660092234611511), ('hunter', 0.5652596950531006), ('obiden', 0.5496266484260559), ('glazed', 0.5360967516899109)]\n"
          ]
        }
      ],
      "source": [
        "# Cell 9: Model 3 — Word2Vec embeddings + Logistic Regression (optimised) \n",
        "\n",
        "\n",
        "!pip install -q gensim\n",
        "from gensim.models import Word2Vec\n",
        "from gensim.utils import simple_preprocess\n",
        "from sklearn.feature_extraction.text import TfidfVectorizer\n",
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.metrics import classification_report, f1_score, accuracy_score, confusion_matrix, ConfusionMatrixDisplay\n",
        "import numpy as np\n",
        "\n",
        "train_tokens = [simple_preprocess(t) for t in X_train]\n",
        "test_tokens  = [simple_preprocess(t) for t in X_test]\n",
        "\n",
        "print('Training Word2Vec (skip-gram)...')\n",
        "w2v = Word2Vec(\n",
        "    sentences=train_tokens,\n",
        "    vector_size=300,\n",
        "    window=7,\n",
        "    min_count=3,\n",
        "    workers=4,\n",
        "    sg=1,\n",
        "    epochs=10\n",
        ")\n",
        "print(f'Vocabulary: {len(w2v.wv):,} words')\n",
        "\n",
        "\n",
        "tfidf_w2v  = TfidfVectorizer(max_features=80000, min_df=3)\n",
        "tfidf_w2v.fit(X_train)\n",
        "tfidf_vocab = tfidf_w2v.vocabulary_\n",
        "idf_vals    = tfidf_w2v._tfidf.idf_\n",
        "\n",
        "def comment_to_vector_tfidf(tokens, model, vocab, idf, size=300):\n",
        "    \"\"\"Mean-pool Word2Vec vectors weighted by TF-IDF IDF scores.\"\"\"\n",
        "    vecs, weights = [], []\n",
        "    for w in tokens:\n",
        "        if w in model.wv and w in vocab:\n",
        "            vecs.append(model.wv[w])\n",
        "            weights.append(idf[vocab[w]])\n",
        "    if not vecs:\n",
        "        return np.zeros(size)\n",
        "    weights = np.array(weights) / np.sum(weights)\n",
        "    return np.average(vecs, axis=0, weights=weights)\n",
        "\n",
        "print('Vectorising train set...')\n",
        "X_train_w2v = np.array([comment_to_vector_tfidf(t, w2v, tfidf_vocab, idf_vals) for t in train_tokens])\n",
        "print('Vectorising test set...')\n",
        "X_test_w2v  = np.array([comment_to_vector_tfidf(t, w2v, tfidf_vocab, idf_vals) for t in test_tokens])\n",
        "\n",
        "lr_w2v = LogisticRegression(class_weight='balanced', max_iter=1000, C=2.0)\n",
        "lr_w2v.fit(X_train_w2v, y_train)\n",
        "y_pred_w2v = lr_w2v.predict(X_test_w2v)\n",
        "\n",
        "acc_w2v = accuracy_score(y_test, y_pred_w2v)\n",
        "f1_w2v  = f1_score(y_test, y_pred_w2v, average='macro')\n",
        "\n",
        "print('\\n=== Model 3: Word2Vec skip-gram + TF-IDF weighted LR (optimised) ===')\n",
        "print(classification_report(y_test, y_pred_w2v, target_names=TARGET_NAMES))\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(6, 5))\n",
        "ConfusionMatrixDisplay(confusion_matrix(y_test, y_pred_w2v), display_labels=TARGET_NAMES).plot(\n",
        "    ax=ax, colorbar=False, cmap='Purples')\n",
        "ax.set_title(f'Word2Vec + LR (optimised)\\nAcc={acc_w2v:.3f}  F1={f1_w2v:.3f}', fontweight='bold')\n",
        "plt.tight_layout(); plt.savefig('cm_word2vec.png', dpi=150, bbox_inches='tight'); plt.show()\n",
        "\n",
        "print('\\n=== Nearest to \"trump\" ===')\n",
        "if 'trump' in w2v.wv: print(w2v.wv.most_similar('trump', topn=8))\n",
        "print('\\n=== Nearest to \"biden\" ===')\n",
        "if 'biden' in w2v.wv: print(w2v.wv.most_similar('biden', topn=8))"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "MV9Ep1mKE8Vi",
      "metadata": {
        "id": "MV9Ep1mKE8Vi"
      },
      "source": [
        "**Cell 10: Model 4 — DistilBERT fine-tuned (with per-epoch eval)**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "WCRg4ZZr1bPh",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000,
          "referenced_widgets": [
            "20b2f026187a4eb0a8163892561a6aec",
            "6d141bf0427845dd9176cdb1e5be3c51",
            "79b85e2ec4dd4300b685f14b85baec89",
            "7f46da9ce1b14598b8ce97e80ffcc895",
            "39cd6a728ff1492d868ac56ad79b5597",
            "8aef5b3cf3254f00a1ea232329d3dea6",
            "6dadf768e0374720acbefcbb41fc97f5",
            "45210055128f4d94a525888ae6ac14d5",
            "5dfb0f739d304191acd3d56a5ae9f42e",
            "90508926a5e8411485057306992ddd16",
            "fee0aab961df4a6cb9cac50eee902401",
            "c745f29d298547b9abea1b58c6521a4a",
            "826972140eff4e46930cf41dab38228a",
            "c8b3b51d4c814c33b5429957e9ef4a30",
            "315f07aebe9a4c548dfb5c97ddf231d2",
            "742c22c8b05d4aeca3b74bdbe952f239",
            "7214f63e9b40410f99117e9cea7a15df",
            "b0ea89429a6e4ca999f8f7491c95a8ce",
            "d0d213c73d52433b9cd63a864e36edca",
            "17759d90d49a407287a76849d87db50d",
            "219eac19d59a47b1b6ace79000e5822c",
            "cb05c044c287480eb17f19672de95300",
            "9196c81f390f450990b1b18ba64e5e0c",
            "fa3a390ef3b3442aaf659ef76b81e0d8",
            "148217a781b1405dae8319f61920e153",
            "b78aeb03ff9e4d249da14c7e97429253",
            "8eb0578c68c048c387a9e6b5f5826dff",
            "06dcb6f236a44b10b22d67fb56f295eb",
            "26462996b4584934a37d20a2b695f7bd",
            "69756039f855413b97c133c0e61d8ffd",
            "4df173a6406e4e17ae67524c48addf2c",
            "ee7bd5ab4c4c43b891e9feb1f70ea2b3",
            "8c7c7dea13764db5980faa0aa7131ed2",
            "089a3536d2184d90b741e5b602c0a047",
            "2ae7adebd0ae466da0fcd2cf43cb9fbd",
            "4ed2b1136adf432eb3665be0761424b5",
            "85182142acec43859a4ed2a7a2962ad4",
            "f0f38c6536834344a6bd9bdfcfc31f09",
            "97751a8adb1d4c75a18be73c74b0bdd6",
            "0ddd416a55a64c9da25450db6c9145b3",
            "716181b6c70e422797fd35ce4d1aa041",
            "55177d72b6c143f0a2d53c5386df48ca",
            "6e061f34446047e991990888052d84a0",
            "7b33c112a64e46d59c6aec1436f54da0",
            "c998997b73934e249f00e4ebf9ce182f",
            "ecb23ee6eeed41feba522f27f3613af0",
            "62b95035cbd242fe828741c93384f768",
            "fdd2c5d394d24f41ae44d88d57c09e9d",
            "5d3a701cde9b4251a59eed82dd89b3a4",
            "66bb5d94f09c4365a95b0b5580d57aaa",
            "0cb5ee3ed56b45f7b04c46ce971c72f0",
            "fd8004950aa5444796794d874c9de3ed",
            "081268dc65e94c5aa33fb71977bd7efc",
            "a7790182bc5546b68b57d3d05da255c8",
            "508fdbb60f5d419c93e07fc82add085a",
            "865113db0cba48f5a37f757a9debfc2e",
            "592c5cc109de4300bf63d15df003085a",
            "d5b7e0589c7c47389a902eee6802e241",
            "04be942dd1924633b01594c5d1e3666b",
            "6b30d117e48844eb8afe9ac39d1ca4d7",
            "99669136ce0a429ebce6cb841d11f206",
            "5b8de7cb42eb4afbb4df181fff80febd",
            "60885a5568f54fffb1dd8643c8213f6f",
            "82498544127d4bb29585f3ab65c9c66d",
            "0a3b99aaff9c42cfac553690a740a1c8",
            "3106bc4951854beaa585ce3685c6da86"
          ]
        },
        "id": "WCRg4ZZr1bPh",
        "outputId": "52e0591c-42d1-4040-e392-a77ef732fdd3"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Device: cuda\n",
            "BERT train size: 100,000\n",
            "label\n",
            "left     50000\n",
            "right    50000\n",
            "Name: count, dtype: int64\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.12/dist-packages/huggingface_hub/utils/_auth.py:93: UserWarning: \n",
            "The secret `HF_TOKEN` does not exist in your Colab secrets.\n",
            "To authenticate with the Hugging Face Hub, create a token in your settings tab (https://huggingface.co/settings/tokens), set it as secret in your Google Colab and restart your session.\n",
            "You will be able to reuse this secret in all of your notebooks.\n",
            "Please note that authentication is recommended but still optional to access public models or datasets.\n",
            "  warnings.warn(\n",
            "Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.\n",
            "WARNING:huggingface_hub.utils._http:Warning: You are sending unauthenticated requests to the HF Hub. Please set a HF_TOKEN to enable higher rate limits and faster downloads.\n"
          ]
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "20b2f026187a4eb0a8163892561a6aec",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "tokenizer_config.json:   0%|          | 0.00/48.0 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "c745f29d298547b9abea1b58c6521a4a",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "vocab.txt: 0.00B [00:00, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "9196c81f390f450990b1b18ba64e5e0c",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "tokenizer.json: 0.00B [00:00, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "089a3536d2184d90b741e5b602c0a047",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "config.json:   0%|          | 0.00/483 [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "c998997b73934e249f00e4ebf9ce182f",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "model.safetensors:   0%|          | 0.00/268M [00:00<?, ?B/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "data": {
            "application/vnd.jupyter.widget-view+json": {
              "model_id": "865113db0cba48f5a37f757a9debfc2e",
              "version_major": 2,
              "version_minor": 0
            },
            "text/plain": [
              "Loading weights:   0%|          | 0/100 [00:00<?, ?it/s]"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "DistilBertForSequenceClassification LOAD REPORT from: distilbert-base-uncased\n",
            "Key                     | Status     | \n",
            "------------------------+------------+-\n",
            "vocab_projector.bias    | UNEXPECTED | \n",
            "vocab_transform.bias    | UNEXPECTED | \n",
            "vocab_layer_norm.weight | UNEXPECTED | \n",
            "vocab_layer_norm.bias   | UNEXPECTED | \n",
            "vocab_transform.weight  | UNEXPECTED | \n",
            "classifier.bias         | MISSING    | \n",
            "pre_classifier.bias     | MISSING    | \n",
            "classifier.weight       | MISSING    | \n",
            "pre_classifier.weight   | MISSING    | \n",
            "\n",
            "Notes:\n",
            "- UNEXPECTED\t:can be ignored when loading from different task/architecture; not ok if you expect identical arch.\n",
            "- MISSING\t:those params were newly initialized because missing from the checkpoint. Consider training on your downstream task.\n",
            "Epoch 1/7: 100%|██████████| 1563/1563 [16:49<00:00,  1.55it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 1/7 — train loss: 0.6211  |  test acc: 0.7023  |  test F1: 0.7023\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Epoch 2/7: 100%|██████████| 1563/1563 [16:59<00:00,  1.53it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 2/7 — train loss: 0.5205  |  test acc: 0.7200  |  test F1: 0.7199\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Epoch 3/7: 100%|██████████| 1563/1563 [16:58<00:00,  1.54it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 3/7 — train loss: 0.4353  |  test acc: 0.7170  |  test F1: 0.7168\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Epoch 4/7: 100%|██████████| 1563/1563 [16:57<00:00,  1.54it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 4/7 — train loss: 0.3506  |  test acc: 0.7257  |  test F1: 0.7246\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Epoch 5/7: 100%|██████████| 1563/1563 [16:58<00:00,  1.53it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 5/7 — train loss: 0.2817  |  test acc: 0.7221  |  test F1: 0.7217\n"
          ]
        },
        {
          "name": "stderr",
          "output_type": "stream",
          "text": [
            "Epoch 6/7: 100%|██████████| 1563/1563 [16:59<00:00,  1.53it/s]\n"
          ]
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Epoch 6/7 — train loss: 0.2290  |  test acc: 0.7209  |  test F1: 0.7205\n",
            "\n",
            "Early stopping triggered after 6 epochs (no improvement for 2 epochs).\n",
            "\n",
            "Best epoch: 4  |  Best F1: 0.7246\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 800x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "=== Model 4: DistilBERT fine-tuned (best epoch = 4) ===\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        left       0.69      0.73      0.71     27428\n",
            "       right       0.76      0.72      0.74     32880\n",
            "\n",
            "    accuracy                           0.73     60308\n",
            "   macro avg       0.72      0.73      0.72     60308\n",
            "weighted avg       0.73      0.73      0.73     60308\n",
            "\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 600x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# ── Cell 10: Model 4 — DistilBERT fine-tuned (with per-epoch eval)\n",
        "\n",
        "import torch\n",
        "import copy\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "from torch.utils.data import Dataset, DataLoader\n",
        "from transformers import (DistilBertTokenizerFast,\n",
        "                          DistilBertForSequenceClassification,\n",
        "                          get_linear_schedule_with_warmup)\n",
        "from torch.optim import AdamW\n",
        "from tqdm import tqdm\n",
        "from sklearn.metrics import (classification_report, f1_score, accuracy_score,\n",
        "                             confusion_matrix, ConfusionMatrixDisplay)\n",
        "\n",
        "\n",
        "if 'DEVICE' not in globals():\n",
        "    DEVICE = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
        "    print(f'DEVICE auto-detected: {DEVICE}')\n",
        "\n",
        "\n",
        "MODEL_NAME   = 'distilbert-base-uncased'\n",
        "MAX_LEN      = 128\n",
        "BATCH_SIZE   = 64 if DEVICE.type == 'cuda' else 16\n",
        "EPOCHS       = 7\n",
        "LR           = 2e-5\n",
        "N_PER_CLASS  = 50000\n",
        "PATIENCE     = 2\n",
        "\n",
        "print(f'Device: {DEVICE}')\n",
        "\n",
        "\n",
        "bert_df = pd.DataFrame({'text': X_train.values, 'label': y_train.values})\n",
        "bert_train_df = (\n",
        "    bert_df.groupby('label', group_keys=False)\n",
        "    .sample(n=min(N_PER_CLASS, bert_df['label'].value_counts().min()),\n",
        "            random_state=42)\n",
        "    .sample(frac=1, random_state=42)\n",
        "    .reset_index(drop=True)\n",
        ")\n",
        "print(f'BERT train size: {len(bert_train_df):,}')\n",
        "print(bert_train_df['label'].value_counts().sort_index()\n",
        "      .rename({i: n for i, n in enumerate(TARGET_NAMES)}))\n",
        "\n",
        "\n",
        "tokenizer = DistilBertTokenizerFast.from_pretrained(MODEL_NAME)\n",
        "\n",
        "class CommentDataset(Dataset):\n",
        "    def __init__(self, texts, labels):\n",
        "        self.encodings = tokenizer(list(texts), truncation=True, padding=True,\n",
        "                                   max_length=MAX_LEN, return_tensors='pt')\n",
        "        self.labels = torch.tensor(list(labels), dtype=torch.long)\n",
        "    def __len__(self):\n",
        "        return len(self.labels)\n",
        "    def __getitem__(self, idx):\n",
        "        return {'input_ids':      self.encodings['input_ids'][idx],\n",
        "                'attention_mask': self.encodings['attention_mask'][idx],\n",
        "                'labels':         self.labels[idx]}\n",
        "\n",
        "train_ds     = CommentDataset(bert_train_df['text'].values,\n",
        "                              bert_train_df['label'].values)\n",
        "test_ds      = CommentDataset(X_test.values, y_test.values)\n",
        "train_loader = DataLoader(train_ds, batch_size=BATCH_SIZE,\n",
        "                          shuffle=True, num_workers=0)\n",
        "test_loader  = DataLoader(test_ds,  batch_size=BATCH_SIZE, num_workers=0)\n",
        "\n",
        "\n",
        "model = DistilBertForSequenceClassification.from_pretrained(\n",
        "    MODEL_NAME, num_labels=2\n",
        ").to(DEVICE)\n",
        "optimizer    = AdamW(model.parameters(), lr=LR, weight_decay=0.01)\n",
        "total_steps  = len(train_loader) * EPOCHS\n",
        "warmup_steps = int(0.1 * total_steps)\n",
        "scheduler    = get_linear_schedule_with_warmup(\n",
        "    optimizer,\n",
        "    num_warmup_steps=warmup_steps,\n",
        "    num_training_steps=total_steps\n",
        ")\n",
        "\n",
        "\n",
        "def evaluate(model, loader):\n",
        "    model.eval()\n",
        "    preds, labels = [], []\n",
        "    with torch.no_grad():\n",
        "        for batch in loader:\n",
        "            out = model(\n",
        "                input_ids=batch['input_ids'].to(DEVICE),\n",
        "                attention_mask=batch['attention_mask'].to(DEVICE)\n",
        "            )\n",
        "            preds.extend(torch.argmax(out.logits, dim=1).cpu().numpy())\n",
        "            labels.extend(batch['labels'].numpy())\n",
        "    return (accuracy_score(labels, preds),\n",
        "            f1_score(labels, preds, average='macro'),\n",
        "            preds, labels)\n",
        "\n",
        "\n",
        "epoch_losses, val_accs, val_f1s = [], [], []\n",
        "best_f1, best_epoch, patience_counter = -1.0, -1, 0\n",
        "best_state = None\n",
        "\n",
        "for epoch in range(EPOCHS):\n",
        "    model.train()\n",
        "    batch_losses = []\n",
        "    for batch in tqdm(train_loader, desc=f'Epoch {epoch+1}/{EPOCHS}'):\n",
        "        loss = model(\n",
        "            input_ids=batch['input_ids'].to(DEVICE),\n",
        "            attention_mask=batch['attention_mask'].to(DEVICE),\n",
        "            labels=batch['labels'].to(DEVICE)\n",
        "        ).loss\n",
        "        batch_losses.append(loss.item())\n",
        "        loss.backward()\n",
        "        torch.nn.utils.clip_grad_norm_(model.parameters(), 1.0)\n",
        "        optimizer.step()\n",
        "        scheduler.step()\n",
        "        optimizer.zero_grad()\n",
        "\n",
        "    avg_loss = sum(batch_losses) / len(batch_losses)\n",
        "    epoch_losses.append(avg_loss)\n",
        "\n",
        "    acc, f1, _, _ = evaluate(model, test_loader)\n",
        "    val_accs.append(acc)\n",
        "    val_f1s.append(f1)\n",
        "    print(f'Epoch {epoch+1}/{EPOCHS} — train loss: {avg_loss:.4f}  |  '\n",
        "          f'test acc: {acc:.4f}  |  test F1: {f1:.4f}')\n",
        "\n",
        "    if f1 > best_f1:\n",
        "        best_f1, best_epoch = f1, epoch + 1\n",
        "        best_state = copy.deepcopy(model.state_dict())\n",
        "        patience_counter = 0\n",
        "    else:\n",
        "        patience_counter += 1\n",
        "        if patience_counter >= PATIENCE:\n",
        "            print(f'\\nEarly stopping triggered after {epoch+1} epochs '\n",
        "                  f'(no improvement for {PATIENCE} epochs).')\n",
        "            break\n",
        "\n",
        "print(f'\\nBest epoch: {best_epoch}  |  Best F1: {best_f1:.4f}')\n",
        "\n",
        "\n",
        "model.load_state_dict(best_state)\n",
        "\n",
        "\n",
        "xs = range(1, len(epoch_losses) + 1)\n",
        "fig, ax1 = plt.subplots(figsize=(8, 5))\n",
        "\n",
        "ax1.plot(xs, epoch_losses, marker='o', color='#E85C4C',\n",
        "         linewidth=2, label='Train loss')\n",
        "ax1.set_xlabel('Epoch')\n",
        "ax1.set_ylabel('Train loss', color='#E85C4C')\n",
        "ax1.tick_params(axis='y', labelcolor='#E85C4C')\n",
        "ax1.set_xticks(list(xs))\n",
        "\n",
        "ax2 = ax1.twinx()\n",
        "ax2.plot(xs, val_f1s, marker='s', color='#4C9BE8',\n",
        "         linewidth=2, label='Test F1 macro')\n",
        "ax2.set_ylabel('Test F1 macro', color='#4C9BE8')\n",
        "ax2.tick_params(axis='y', labelcolor='#4C9BE8')\n",
        "ax2.axvline(best_epoch, color='grey', linestyle='--', alpha=0.5,\n",
        "            label=f'Best epoch ({best_epoch})')\n",
        "ax2.legend(loc='lower right')\n",
        "\n",
        "plt.title('DistilBERT — train loss vs test F1 per epoch', fontweight='bold')\n",
        "plt.tight_layout()\n",
        "plt.savefig('distilbert_loss.png', dpi=150, bbox_inches='tight')\n",
        "plt.show()\n",
        "\n",
        "\n",
        "acc_bert, f1_bert, all_preds, all_labels_bert = evaluate(model, test_loader)\n",
        "\n",
        "print(f'\\n=== Model 4: DistilBERT fine-tuned (best epoch = {best_epoch}) ===')\n",
        "print(classification_report(all_labels_bert, all_preds,\n",
        "                            target_names=TARGET_NAMES))\n",
        "\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(6, 5))\n",
        "ConfusionMatrixDisplay(\n",
        "    confusion_matrix(all_labels_bert, all_preds),\n",
        "    display_labels=TARGET_NAMES\n",
        ").plot(ax=ax, colorbar=False, cmap='Reds')\n",
        "\n",
        "for text in ax.texts:\n",
        "    try:\n",
        "        val = int(float(text.get_text()))\n",
        "        text.set_text(f'{val:,}')\n",
        "    except ValueError:\n",
        "        pass\n",
        "\n",
        "ax.set_title(f'DistilBERT fine-tuned (best epoch {best_epoch})\\n'\n",
        "             f'Acc={acc_bert:.3f}  F1={f1_bert:.3f}',\n",
        "             fontweight='bold')\n",
        "plt.tight_layout()\n",
        "plt.savefig('cm_distilbert.png', dpi=150, bbox_inches='tight')\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "viTB0BYcFIfn",
      "metadata": {
        "id": "viTB0BYcFIfn"
      },
      "source": [
        "**Cell 11: Model 5 — GPT-4o few-shot prompting**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "lp3IPBZq1bPi",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lp3IPBZq1bPi",
        "outputId": "46a848c2-7527-40fc-f4e0-e53c879dfebb"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "GPT sample: 102 comments ({0: 51, 1: 51})\n",
            "Running zero-shot...\n",
            "Running few-shot...\n",
            "\n",
            "GPT Zero-shot (n=102): Acc=0.500  F1=0.500\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        left       0.50      0.51      0.50        51\n",
            "       right       0.50      0.49      0.50        51\n",
            "\n",
            "    accuracy                           0.50       102\n",
            "   macro avg       0.50      0.50      0.50       102\n",
            "weighted avg       0.50      0.50      0.50       102\n",
            "\n",
            "\n",
            "GPT Few-shot (n=102): Acc=0.520  F1=0.519\n",
            "              precision    recall  f1-score   support\n",
            "\n",
            "        left       0.52      0.49      0.51        51\n",
            "       right       0.52      0.55      0.53        51\n",
            "\n",
            "    accuracy                           0.52       102\n",
            "   macro avg       0.52      0.52      0.52       102\n",
            "weighted avg       0.52      0.52      0.52       102\n",
            "\n"
          ]
        }
      ],
      "source": [
        "# Cell 11: Model 5 — GPT-4o few-shot prompting\n",
        "\n",
        "\n",
        "import re, time, pandas as pd\n",
        "from sklearn.metrics import accuracy_score, f1_score, classification_report\n",
        "from openai import OpenAI\n",
        "import os\n",
        "import sys\n",
        "\n",
        "\n",
        "if 'df_model' not in globals():\n",
        "    if 'df' not in globals():\n",
        "\n",
        "        if 'DATA_PATH' not in globals():\n",
        "            IN_COLAB = 'google.colab' in sys.modules\n",
        "            if IN_COLAB:\n",
        "                DATA_PATH = '/content/drive/MyDrive/NLP/youtube_comments_clean.csv'\n",
        "            else:\n",
        "                DATA_PATH = 'youtube_comments_clean.csv'\n",
        "            print(f'Re-defined DATA_PATH: {DATA_PATH}')\n",
        "        df = pd.read_csv(DATA_PATH)\n",
        "    def _clean(text):\n",
        "        if not isinstance(text, str): return ''\n",
        "        text = text.lower()\n",
        "        text = re.sub(r'http\\S+|www\\S+', ' ', text)\n",
        "        text = re.sub(r'[^ȴ\\s]', ' ', text)\n",
        "        return re.sub(r'\\s+', ' ', text).strip()\n",
        "    df.loc[df['channel_name'] == 'CNN', 'channel_bias'] = 'left'\n",
        "    df_en = df[df['language'] == 'en'].copy().reset_index(drop=True)\n",
        "    df_en['text_clean'] = df_en['text'].apply(_clean)\n",
        "    df_en = df_en[df_en['text_clean'].str.len() > 0].reset_index(drop=True)\n",
        "    LABEL_MAP    = {'left': 0, 'right': 1}\n",
        "    TARGET_NAMES = ['left', 'right']\n",
        "    df_model = df_en[df_en['channel_bias'].isin(LABEL_MAP)].copy()\n",
        "    df_model['label'] = df_model['channel_bias'].map(LABEL_MAP)\n",
        "    print(f'Rebuilt df_model: {len(df_model):,} rows')\n",
        "\n",
        "from google.colab import userdata\n",
        "OPENAI_API_KEY = userdata.get('OPENAI_API_KEY')\n",
        "client = OpenAI(api_key=OPENAI_API_KEY)\n",
        "\n",
        "\n",
        "SYSTEM_ZERO = \"\"\"You are an expert in US political media.\n",
        "Classify which type of YouTube channel this comment was most likely posted on.\n",
        "Reply with exactly one of: left, right.\"\"\"\n",
        "\n",
        "FEW_SHOT_EXAMPLES = [\n",
        "    ('tax the billionaires and fund public schools now', 'left'),\n",
        "    ('we need stronger gun control laws to protect our kids', 'left'),\n",
        "    ('the corporate media ignores what really matters to working people', 'left'),\n",
        "    ('project 2025 is a real threat to american democracy', 'left'),\n",
        "    ('medicare for all is the only humane solution', 'left'),\n",
        "    ('the mainstream media is lying about the border crisis', 'right'),\n",
        "    ('small government and low taxes is what made america great', 'right'),\n",
        "    ('the democrats want open borders and socialism for everyone', 'right'),\n",
        "    ('god bless president trump and the usa', 'right'),\n",
        "    ('the second amendment shall not be infringed period', 'right'),\n",
        "]\n",
        "few_shot_str = '\\n'.join([f'Comment: \"{ex}\"\\nLabel: {lbl}'\n",
        "                          for ex, lbl in FEW_SHOT_EXAMPLES])\n",
        "SYSTEM_FEW = SYSTEM_ZERO + '\\n\\nExamples:\\n' + few_shot_str\n",
        "\n",
        "def parse_label(answer: str) -> int:\n",
        "    a = answer.strip().lower()\n",
        "    if 'right' in a: return 1\n",
        "    if 'left'  in a: return 0\n",
        "    return -1\n",
        "\n",
        "def gpt_classify(text, system_prompt, model='gpt-4o-mini') -> int:\n",
        "    for attempt in range(3):\n",
        "        try:\n",
        "            resp = client.chat.completions.create(\n",
        "                model=model,\n",
        "                messages=[\n",
        "                    {'role': 'system', 'content': system_prompt},\n",
        "                    {'role': 'user',   'content': f'Comment: \"{text[:400]}\"'}\n",
        "                ],\n",
        "                max_tokens=10, temperature=0\n",
        "            )\n",
        "            return parse_label(resp.choices[0].message.content)\n",
        "        except Exception:\n",
        "            time.sleep(2 ** attempt)\n",
        "    return -1\n",
        "\n",
        "\n",
        "sample = pd.concat([\n",
        "    g.sample(min(len(g), 51), random_state=42)\n",
        "    for _, g in df_model.groupby('label', group_keys=False)\n",
        "]).sample(frac=1, random_state=42).reset_index(drop=True)\n",
        "\n",
        "print(f'GPT sample: {len(sample)} comments ({sample[\"label\"].value_counts().to_dict()})')\n",
        "print('Running zero-shot...')\n",
        "sample['gpt_zero'] = sample['text_clean'].apply(lambda t: gpt_classify(t, SYSTEM_ZERO))\n",
        "print('Running few-shot...')\n",
        "sample['gpt_few']  = sample['text_clean'].apply(lambda t: gpt_classify(t, SYSTEM_FEW))\n",
        "\n",
        "\n",
        "for col, name in [('gpt_zero', 'Zero-shot'), ('gpt_few', 'Few-shot')]:\n",
        "    valid = sample[sample[col] != -1]\n",
        "    acc = accuracy_score(valid['label'], valid[col])\n",
        "    f1  = f1_score(valid['label'], valid[col], average='macro')\n",
        "    print(f'\\nGPT {name} (n={len(valid)}): Acc={acc:.3f}  F1={f1:.3f}')\n",
        "    print(classification_report(valid['label'], valid[col], target_names=TARGET_NAMES))\n",
        "\n",
        "acc_gpt_zero = accuracy_score(sample[sample['gpt_zero']!=-1]['label'],\n",
        "                               sample[sample['gpt_zero']!=-1]['gpt_zero'])\n",
        "acc_gpt_few  = accuracy_score(sample[sample['gpt_few']!=-1]['label'],\n",
        "                               sample[sample['gpt_few']!=-1]['gpt_few'])\n",
        "f1_gpt_zero  = f1_score(sample[sample['gpt_zero']!=-1]['label'],\n",
        "                         sample[sample['gpt_zero']!=-1]['gpt_zero'], average='macro')\n",
        "f1_gpt_few   = f1_score(sample[sample['gpt_few']!=-1]['label'],\n",
        "                         sample[sample['gpt_few']!=-1]['gpt_few'],  average='macro')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "Q-jT3JdN1bPi",
      "metadata": {
        "id": "Q-jT3JdN1bPi"
      },
      "source": [
        "---\n",
        "## Results & Analysis\n"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "lo457GXiFL9e",
      "metadata": {
        "id": "lo457GXiFL9e"
      },
      "source": [
        "**Cell 12: Model Comparison — Results**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "RhCmcDg41bPi",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 803
        },
        "id": "RhCmcDg41bPi",
        "outputId": "f773c3c2-825b-4acc-d5ee-5b392ecfc87d"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "=== Model Comparison — 2-class Political Ideology Classification ===\n",
            "                         Model  Accuracy  F1 macro\n",
            "            ComplementNB + BoW     0.707     0.706\n",
            "  LinearSVC + TF-IDF word+char     0.719     0.718\n",
            "Word2Vec skip-gram + TF-IDF LR     0.664     0.664\n",
            "         DistilBERT fine-tuned     0.726     0.725\n",
            "              GPT-4o zero-shot     0.500     0.500\n",
            "               GPT-4o few-shot     0.520     0.519\n"
          ]
        },
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1100x500 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Saved: model_comparison.png\n",
            "\n",
            "=== Cross-channel Generalisation (LinearSVC) ===\n",
            "Condition                                  F1 macro\n",
            "----------------------------------------------------\n",
            "Within-channel (standard split)               0.718\n",
            "Cross-channel — baseline                      0.592\n",
            "Cross-channel — debiased                      0.585\n",
            "Channel identity effect                       0.126\n",
            "Debiasing recovery                           -0.007\n"
          ]
        }
      ],
      "source": [
        "#  Cell 12: Model Comparison — Results \n",
        "\n",
        "import matplotlib.pyplot as plt\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "from sklearn.metrics import accuracy_score, f1_score\n",
        "\n",
        "def scores(y_true, y_pred):\n",
        "    return {\n",
        "        'Accuracy': round(accuracy_score(y_true, y_pred), 3),\n",
        "        'F1 macro': round(f1_score(y_true, y_pred, average='macro'), 3),\n",
        "    }\n",
        "\n",
        "results = [\n",
        "    {'Model': 'ComplementNB + BoW',             **scores(y_test, y_pred_nb)},\n",
        "    {'Model': 'LinearSVC + TF-IDF word+char',   **scores(y_test, y_pred_lr)},\n",
        "    {'Model': 'Word2Vec skip-gram + TF-IDF LR', **scores(y_test, y_pred_w2v)},\n",
        "    {'Model': 'DistilBERT fine-tuned',           **scores(all_labels_bert, all_preds)},\n",
        "    {'Model': 'GPT-4o zero-shot',\n",
        "     'Accuracy': round(acc_gpt_zero, 3), 'F1 macro': round(f1_gpt_zero, 3)},\n",
        "    {'Model': 'GPT-4o few-shot',\n",
        "     'Accuracy': round(acc_gpt_few, 3),  'F1 macro': round(f1_gpt_few, 3)},\n",
        "]\n",
        "\n",
        "results_df = pd.DataFrame(results)\n",
        "print('=== Model Comparison — 2-class Political Ideology Classification ===')\n",
        "print(results_df.to_string(index=False))\n",
        "\n",
        "\n",
        "fig, ax = plt.subplots(figsize=(11, 5))\n",
        "x     = np.arange(len(results))\n",
        "width = 0.35\n",
        "\n",
        "acc_vals = [r['Accuracy'] for r in results]\n",
        "f1_vals  = [r['F1 macro'] for r in results]\n",
        "\n",
        "bars1 = ax.bar(x - width/2, acc_vals, width, label='Accuracy',\n",
        "               color='#4C9BE8', alpha=0.85, edgecolor='white')\n",
        "bars2 = ax.bar(x + width/2, f1_vals,  width, label='F1 macro',\n",
        "               color='#E85C4C', alpha=0.85, edgecolor='white')\n",
        "\n",
        "ax.set_ylabel('Score')\n",
        "ax.set_title('Model Comparison — 2-class Political Ideology Classification',\n",
        "             fontsize=13, fontweight='bold')\n",
        "ax.set_xticks(x)\n",
        "ax.set_xticklabels([r['Model'] for r in results],\n",
        "                   rotation=18, ha='right', fontsize=9)\n",
        "ax.set_ylim(0, 1.0)\n",
        "ax.axhline(0.5, color='grey', linestyle='--', linewidth=1,\n",
        "           label='Random baseline (2-class)')\n",
        "ax.legend(fontsize=9)\n",
        "ax.grid(axis='y', alpha=0.4)\n",
        "\n",
        "for bar in list(bars1) + list(bars2):\n",
        "    ax.text(bar.get_x() + bar.get_width() / 2, bar.get_height() + 0.012,\n",
        "            f'{bar.get_height():.3f}', ha='center', va='bottom', fontsize=8)\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.savefig('model_comparison.png', dpi=150, bbox_inches='tight')\n",
        "plt.show()\n",
        "print('Saved: model_comparison.png')\n",
        "\n",
        "\n",
        "print('\\n=== Cross-channel Generalisation (LinearSVC) ===')\n",
        "print(f'{\"Condition\":<40} {\"F1 macro\":>10}')\n",
        "print('-' * 52)\n",
        "print(f'{\"Within-channel (standard split)\":<40} {f1_lr:>10.3f}')\n",
        "print(f'{\"Cross-channel — baseline\":<40} {f1_cross_baseline:>10.3f}')\n",
        "print(f'{\"Cross-channel — debiased\":<40} {f1_cross_debiased:>10.3f}')\n",
        "print(f'{\"Channel identity effect\":<40} {f1_lr - f1_cross_baseline:>10.3f}')\n",
        "print(f'{\"Debiasing recovery\":<40} {f1_cross_debiased - f1_cross_baseline:>+10.3f}')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "He210-VjFT6F",
      "metadata": {
        "id": "He210-VjFT6F"
      },
      "source": [
        "**Cell 13: Error Analysis**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "tWnDy_lu1bPi",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tWnDy_lu1bPi",
        "outputId": "59a192b8-c46c-4ed2-9e51-b2a56b7ada27"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "LR vs BERT disagreement total: 14,256\n",
            "  BERT correct, LR wrong: 7,329\n",
            "  LR correct, BERT wrong: 6,927\n",
            "  Symmetry ratio:         1.058\n",
            "  → Symmetric disagreement supports the ensemble extension in Conclusion.\n",
            "Comments where ALL models fail: 7,532\n",
            "\n",
            "Sample hard examples:\n",
            "  [true=right] americans are sick of these crooks\n",
            "  [true=left] it s all fun games swimming in the pool with fdjt until fdjt decides to poop in it\n",
            "  [true=right] so much begging from the usa\n",
            "  [true=left] abolish government we all anarchy\n",
            "  [true=left] at least obama knew how to treat cancer\n",
            "  [true=right] pain and simple corruption\n",
            "  [true=left] why isn t the fbi utah state police etc not trying to get to the truth unbelievable\n",
            "  [true=right] not all americans are maga so deal with it\n",
            "\n",
            "BERT correct, LR wrong: 7,329 (contextual understanding helps)\n",
            "  [true=right] when you rob a of their land there will be consequences\n",
            "  [true=left] because pistol doesn t have any answers\n",
            "  [true=left] but it is the senate that will never move\n",
            "  [true=left] now i trump is one of the clients\n",
            "  [true=right] the country was never safe mostly due to decades of racism and apartheid\n",
            "\n",
            "LR correct, BERT wrong: 6,927 (keywords suffice)\n",
            "  [true=right] send the marines in to chicago and get michael bay to film the fuck out of it make it look the end of a transformers mov\n",
            "  [true=left] can any country do a blockade if they have the power and confiscate the goods sounds piracy\n",
            "  [true=right] there really is something strange happening and not the usual strange it s something different the left will that it s t\n",
            "  [true=right] he has enemies that are not away so better get your act together stop being a bully these attacks should be sending you \n",
            "  [true=right] 9 30 this makes no sense\n"
          ]
        }
      ],
      "source": [
        "# Cell 13: Error Analysis — Hard Examples \n",
        "\n",
        "label_map = {i: n for i, n in enumerate(TARGET_NAMES)}\n",
        "\n",
        "test_df = pd.DataFrame({\n",
        "    'text':       X_test.values,\n",
        "    'true':       [label_map[i] for i in y_test.values],\n",
        "    'pred_nb':    [label_map[i] for i in y_pred_nb],\n",
        "    'pred_lr':    [label_map[i] for i in y_pred_lr],\n",
        "    'pred_bert':  [label_map[i] for i in all_preds],\n",
        "})\n",
        "test_df['true_label'] = y_test.values\n",
        "test_df['bert_label'] = all_preds\n",
        "\n",
        "\n",
        "\n",
        "n_lr_bert_disagree = (test_df['pred_lr'] != test_df['pred_bert']).sum()\n",
        "n_bert_wins_lr_loses = ((test_df['pred_bert'] == test_df['true']) &\n",
        "                       (test_df['pred_lr']   != test_df['true'])).sum()\n",
        "n_lr_wins_bert_loses = ((test_df['pred_lr']   == test_df['true']) &\n",
        "                       (test_df['pred_bert'] != test_df['true'])).sum()\n",
        "print(f'LR vs BERT disagreement total: {n_lr_bert_disagree:,}')\n",
        "print(f'  BERT correct, LR wrong: {n_bert_wins_lr_loses:,}')\n",
        "print(f'  LR correct, BERT wrong: {n_lr_wins_bert_loses:,}')\n",
        "print(f'  Symmetry ratio:         {n_bert_wins_lr_loses / max(n_lr_wins_bert_loses, 1):.3f}')\n",
        "print(f'  → Symmetric disagreement supports the ensemble extension in Conclusion.')\n",
        "\n",
        "\n",
        "all_wrong = test_df[\n",
        "    (test_df['pred_nb']   != test_df['true']) &\n",
        "    (test_df['pred_lr']   != test_df['true']) &\n",
        "    (test_df['pred_bert'] != test_df['true'])\n",
        "]\n",
        "print(f'Comments where ALL models fail: {len(all_wrong):,}')\n",
        "print('\\nSample hard examples:')\n",
        "for _, row in all_wrong.sample(min(8, len(all_wrong)), random_state=42).iterrows():\n",
        "    print(f'  [true={row[\"true\"]}] {row[\"text\"][:120]}')\n",
        "\n",
        "\n",
        "bert_wins = test_df[\n",
        "    (test_df['pred_bert'] == test_df['true']) &\n",
        "    (test_df['pred_lr']   != test_df['true'])\n",
        "]\n",
        "print(f'\\nBERT correct, LR wrong: {len(bert_wins):,} (contextual understanding helps)')\n",
        "for _, row in bert_wins.sample(min(5, len(bert_wins)), random_state=1).iterrows():\n",
        "    print(f'  [true={row[\"true\"]}] {row[\"text\"][:120]}')\n",
        "\n",
        "\n",
        "lr_wins = test_df[\n",
        "    (test_df['pred_lr']   == test_df['true']) &\n",
        "    (test_df['pred_bert'] != test_df['true'])\n",
        "]\n",
        "print(f'\\nLR correct, BERT wrong: {len(lr_wins):,} (keywords suffice)')\n",
        "for _, row in lr_wins.sample(min(5, len(lr_wins)), random_state=2).iterrows():\n",
        "    print(f'  [true={row[\"true\"]}] {row[\"text\"][:120]}')"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "JhFqKBNzFanP",
      "metadata": {
        "id": "JhFqKBNzFanP"
      },
      "source": [
        "**Cell 14: LDA Topic Modelling — Bonus qualitative analysis**"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "lK9PQNGx1bPi",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lK9PQNGx1bPi",
        "outputId": "19e8dd19-47c0-46d0-8007-cc2aad25270c"
      },
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "=== LDA topics — LEFT ===\n",
            "  Topic 1: 0.016*\"trump\" + 0.013*\"oil\" + 0.012*\"money\" + 0.010*\"war\" + 0.008*\"iran\" + 0.006*\"pay\" + 0.006*\"american\" + 0.006*\"get\"\n",
            "  Topic 2: 0.022*\"iran\" + 0.014*\"trump\" + 0.012*\"israel\" + 0.010*\"war\" + 0.009*\"would\" + 0.009*\"even\" + 0.008*\"china\" + 0.007*\"get\"\n",
            "  Topic 3: 0.021*\"trump\" + 0.015*\"israel\" + 0.008*\"iran\" + 0.007*\"states\" + 0.007*\"stop\" + 0.007*\"one\" + 0.007*\"united\" + 0.006*\"america\"\n",
            "  Topic 4: 0.039*\"trump\" + 0.020*\"war\" + 0.011*\"iran\" + 0.009*\"world\" + 0.009*\"news\" + 0.007*\"one\" + 0.007*\"america\" + 0.007*\"military\"\n",
            "  Topic 5: 0.032*\"trump\" + 0.010*\"would\" + 0.009*\"president\" + 0.008*\"never\" + 0.008*\"get\" + 0.008*\"god\" + 0.007*\"one\" + 0.006*\"vote\"\n",
            "\n",
            "=== LDA topics — RIGHT ===\n",
            "  Topic 1: 0.019*\"trump\" + 0.009*\"get\" + 0.009*\"would\" + 0.009*\"democrats\" + 0.007*\"one\" + 0.006*\"left\" + 0.006*\"right\" + 0.005*\"money\"\n",
            "  Topic 2: 0.024*\"iran\" + 0.017*\"trump\" + 0.008*\"america\" + 0.007*\"israel\" + 0.007*\"states\" + 0.006*\"new\" + 0.006*\"country\" + 0.006*\"would\"\n",
            "  Topic 3: 0.010*\"one\" + 0.009*\"war\" + 0.008*\"right\" + 0.008*\"israel\" + 0.006*\"world\" + 0.006*\"american\" + 0.006*\"america\" + 0.006*\"would\"\n",
            "  Topic 4: 0.030*\"god\" + 0.026*\"trump\" + 0.018*\"president\" + 0.015*\"thank\" + 0.014*\"love\" + 0.011*\"bless\" + 0.010*\"man\" + 0.010*\"great\"\n",
            "  Topic 5: 0.007*\"even\" + 0.007*\"would\" + 0.007*\"never\" + 0.006*\"get\" + 0.006*\"one\" + 0.005*\"security\" + 0.005*\"state\" + 0.005*\"government\"\n",
            "\n",
            "Done — compare the topics side-by-side for qualitative discussion in the paper.\n"
          ]
        }
      ],
      "source": [
        "# ── Cell 14: LDA Topic Modelling — Bonus qualitative analysis\n",
        "\n",
        "\n",
        "import gensim.corpora as corpora\n",
        "from gensim.models import LdaMulticore\n",
        "from gensim.utils import simple_preprocess\n",
        "import nltk\n",
        "nltk.download('stopwords', quiet=True)\n",
        "from nltk.corpus import stopwords\n",
        "\n",
        "STOP = set(stopwords.words('english'))\n",
        "NUM_TOPICS = 5\n",
        "\n",
        "def run_lda(texts, n_topics=5, label=''):\n",
        "    tokens = [\n",
        "        [w for w in simple_preprocess(t) if w not in STOP and len(w) > 2]\n",
        "        for t in texts\n",
        "    ]\n",
        "    dictionary = corpora.Dictionary(tokens)\n",
        "    dictionary.filter_extremes(no_below=10, no_above=0.5)\n",
        "    corpus = [dictionary.doc2bow(t) for t in tokens]\n",
        "    lda = LdaMulticore(corpus, num_topics=n_topics, id2word=dictionary,\n",
        "                       passes=5, workers=2, random_state=42)\n",
        "    print(f'\\n=== LDA topics — {label} ===')\n",
        "    for i, topic in lda.print_topics(num_words=8):\n",
        "        print(f'  Topic {i+1}: {topic}')\n",
        "    return lda\n",
        "\n",
        "SAMPLE_N = 30000\n",
        "left_pool  = df_model[df_model['label'] == 0]['text_clean']\n",
        "right_pool = df_model[df_model['label'] == 1]['text_clean']\n",
        "\n",
        "left_sample  = left_pool.sample(min(SAMPLE_N, len(left_pool)),  random_state=42)\n",
        "right_sample = right_pool.sample(min(SAMPLE_N, len(right_pool)), random_state=42)\n",
        "\n",
        "lda_left  = run_lda(left_sample,  label='LEFT')\n",
        "lda_right = run_lda(right_sample, label='RIGHT')\n",
        "\n",
        "print('\\nDone — compare the topics side-by-side for qualitative discussion in the paper.')"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "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": "3.10.11"
    },
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "04be942dd1924633b01594c5d1e3666b": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_0a3b99aaff9c42cfac553690a740a1c8",
            "placeholder": "​",
            "style": "IPY_MODEL_3106bc4951854beaa585ce3685c6da86",
            "value": " 100/100 [00:00&lt;00:00, 483.86it/s, Materializing param=distilbert.transformer.layer.5.sa_layer_norm.weight]"
          }
        },
        "06dcb6f236a44b10b22d67fb56f295eb": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "081268dc65e94c5aa33fb71977bd7efc": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "089a3536d2184d90b741e5b602c0a047": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_2ae7adebd0ae466da0fcd2cf43cb9fbd",
              "IPY_MODEL_4ed2b1136adf432eb3665be0761424b5",
              "IPY_MODEL_85182142acec43859a4ed2a7a2962ad4"
            ],
            "layout": "IPY_MODEL_f0f38c6536834344a6bd9bdfcfc31f09"
          }
        },
        "0a3b99aaff9c42cfac553690a740a1c8": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "0cb5ee3ed56b45f7b04c46ce971c72f0": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "0ddd416a55a64c9da25450db6c9145b3": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "148217a781b1405dae8319f61920e153": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_69756039f855413b97c133c0e61d8ffd",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_4df173a6406e4e17ae67524c48addf2c",
            "value": 1
          }
        },
        "17759d90d49a407287a76849d87db50d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "20b2f026187a4eb0a8163892561a6aec": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_6d141bf0427845dd9176cdb1e5be3c51",
              "IPY_MODEL_79b85e2ec4dd4300b685f14b85baec89",
              "IPY_MODEL_7f46da9ce1b14598b8ce97e80ffcc895"
            ],
            "layout": "IPY_MODEL_39cd6a728ff1492d868ac56ad79b5597"
          }
        },
        "219eac19d59a47b1b6ace79000e5822c": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "26462996b4584934a37d20a2b695f7bd": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "2ae7adebd0ae466da0fcd2cf43cb9fbd": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_97751a8adb1d4c75a18be73c74b0bdd6",
            "placeholder": "​",
            "style": "IPY_MODEL_0ddd416a55a64c9da25450db6c9145b3",
            "value": "config.json: 100%"
          }
        },
        "3106bc4951854beaa585ce3685c6da86": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "315f07aebe9a4c548dfb5c97ddf231d2": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_219eac19d59a47b1b6ace79000e5822c",
            "placeholder": "​",
            "style": "IPY_MODEL_cb05c044c287480eb17f19672de95300",
            "value": " 232k/? [00:00&lt;00:00, 2.72MB/s]"
          }
        },
        "39cd6a728ff1492d868ac56ad79b5597": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "45210055128f4d94a525888ae6ac14d5": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "4df173a6406e4e17ae67524c48addf2c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "4ed2b1136adf432eb3665be0761424b5": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_716181b6c70e422797fd35ce4d1aa041",
            "max": 483,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_55177d72b6c143f0a2d53c5386df48ca",
            "value": 483
          }
        },
        "508fdbb60f5d419c93e07fc82add085a": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "55177d72b6c143f0a2d53c5386df48ca": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "592c5cc109de4300bf63d15df003085a": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_99669136ce0a429ebce6cb841d11f206",
            "placeholder": "​",
            "style": "IPY_MODEL_5b8de7cb42eb4afbb4df181fff80febd",
            "value": "Loading weights: 100%"
          }
        },
        "5b8de7cb42eb4afbb4df181fff80febd": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "5d3a701cde9b4251a59eed82dd89b3a4": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "5dfb0f739d304191acd3d56a5ae9f42e": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "60885a5568f54fffb1dd8643c8213f6f": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "62b95035cbd242fe828741c93384f768": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_fd8004950aa5444796794d874c9de3ed",
            "max": 267954768,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_081268dc65e94c5aa33fb71977bd7efc",
            "value": 267954768
          }
        },
        "66bb5d94f09c4365a95b0b5580d57aaa": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "69756039f855413b97c133c0e61d8ffd": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": "20px"
          }
        },
        "6b30d117e48844eb8afe9ac39d1ca4d7": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6d141bf0427845dd9176cdb1e5be3c51": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_8aef5b3cf3254f00a1ea232329d3dea6",
            "placeholder": "​",
            "style": "IPY_MODEL_6dadf768e0374720acbefcbb41fc97f5",
            "value": "tokenizer_config.json: 100%"
          }
        },
        "6dadf768e0374720acbefcbb41fc97f5": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "6e061f34446047e991990888052d84a0": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "716181b6c70e422797fd35ce4d1aa041": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "7214f63e9b40410f99117e9cea7a15df": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "742c22c8b05d4aeca3b74bdbe952f239": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "79b85e2ec4dd4300b685f14b85baec89": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_45210055128f4d94a525888ae6ac14d5",
            "max": 48,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_5dfb0f739d304191acd3d56a5ae9f42e",
            "value": 48
          }
        },
        "7b33c112a64e46d59c6aec1436f54da0": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "7f46da9ce1b14598b8ce97e80ffcc895": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_90508926a5e8411485057306992ddd16",
            "placeholder": "​",
            "style": "IPY_MODEL_fee0aab961df4a6cb9cac50eee902401",
            "value": " 48.0/48.0 [00:00&lt;00:00, 2.21kB/s]"
          }
        },
        "82498544127d4bb29585f3ab65c9c66d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "ProgressStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "826972140eff4e46930cf41dab38228a": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_7214f63e9b40410f99117e9cea7a15df",
            "placeholder": "​",
            "style": "IPY_MODEL_b0ea89429a6e4ca999f8f7491c95a8ce",
            "value": "vocab.txt: "
          }
        },
        "85182142acec43859a4ed2a7a2962ad4": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6e061f34446047e991990888052d84a0",
            "placeholder": "​",
            "style": "IPY_MODEL_7b33c112a64e46d59c6aec1436f54da0",
            "value": " 483/483 [00:00&lt;00:00, 57.7kB/s]"
          }
        },
        "865113db0cba48f5a37f757a9debfc2e": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_592c5cc109de4300bf63d15df003085a",
              "IPY_MODEL_d5b7e0589c7c47389a902eee6802e241",
              "IPY_MODEL_04be942dd1924633b01594c5d1e3666b"
            ],
            "layout": "IPY_MODEL_6b30d117e48844eb8afe9ac39d1ca4d7"
          }
        },
        "8aef5b3cf3254f00a1ea232329d3dea6": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "8c7c7dea13764db5980faa0aa7131ed2": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "8eb0578c68c048c387a9e6b5f5826dff": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "90508926a5e8411485057306992ddd16": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "9196c81f390f450990b1b18ba64e5e0c": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_fa3a390ef3b3442aaf659ef76b81e0d8",
              "IPY_MODEL_148217a781b1405dae8319f61920e153",
              "IPY_MODEL_b78aeb03ff9e4d249da14c7e97429253"
            ],
            "layout": "IPY_MODEL_8eb0578c68c048c387a9e6b5f5826dff"
          }
        },
        "97751a8adb1d4c75a18be73c74b0bdd6": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "99669136ce0a429ebce6cb841d11f206": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "a7790182bc5546b68b57d3d05da255c8": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b0ea89429a6e4ca999f8f7491c95a8ce": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "b78aeb03ff9e4d249da14c7e97429253": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_ee7bd5ab4c4c43b891e9feb1f70ea2b3",
            "placeholder": "​",
            "style": "IPY_MODEL_8c7c7dea13764db5980faa0aa7131ed2",
            "value": " 466k/? [00:00&lt;00:00, 3.59MB/s]"
          }
        },
        "c745f29d298547b9abea1b58c6521a4a": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_826972140eff4e46930cf41dab38228a",
              "IPY_MODEL_c8b3b51d4c814c33b5429957e9ef4a30",
              "IPY_MODEL_315f07aebe9a4c548dfb5c97ddf231d2"
            ],
            "layout": "IPY_MODEL_742c22c8b05d4aeca3b74bdbe952f239"
          }
        },
        "c8b3b51d4c814c33b5429957e9ef4a30": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_d0d213c73d52433b9cd63a864e36edca",
            "max": 1,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_17759d90d49a407287a76849d87db50d",
            "value": 1
          }
        },
        "c998997b73934e249f00e4ebf9ce182f": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HBoxModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_ecb23ee6eeed41feba522f27f3613af0",
              "IPY_MODEL_62b95035cbd242fe828741c93384f768",
              "IPY_MODEL_fdd2c5d394d24f41ae44d88d57c09e9d"
            ],
            "layout": "IPY_MODEL_5d3a701cde9b4251a59eed82dd89b3a4"
          }
        },
        "cb05c044c287480eb17f19672de95300": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "d0d213c73d52433b9cd63a864e36edca": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": "20px"
          }
        },
        "d5b7e0589c7c47389a902eee6802e241": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "FloatProgressModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_60885a5568f54fffb1dd8643c8213f6f",
            "max": 100,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_82498544127d4bb29585f3ab65c9c66d",
            "value": 100
          }
        },
        "ecb23ee6eeed41feba522f27f3613af0": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_66bb5d94f09c4365a95b0b5580d57aaa",
            "placeholder": "​",
            "style": "IPY_MODEL_0cb5ee3ed56b45f7b04c46ce971c72f0",
            "value": "model.safetensors: 100%"
          }
        },
        "ee7bd5ab4c4c43b891e9feb1f70ea2b3": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f0f38c6536834344a6bd9bdfcfc31f09": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fa3a390ef3b3442aaf659ef76b81e0d8": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_06dcb6f236a44b10b22d67fb56f295eb",
            "placeholder": "​",
            "style": "IPY_MODEL_26462996b4584934a37d20a2b695f7bd",
            "value": "tokenizer.json: "
          }
        },
        "fd8004950aa5444796794d874c9de3ed": {
          "model_module": "@jupyter-widgets/base",
          "model_module_version": "1.2.0",
          "model_name": "LayoutModel",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "fdd2c5d394d24f41ae44d88d57c09e9d": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "HTMLModel",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_a7790182bc5546b68b57d3d05da255c8",
            "placeholder": "​",
            "style": "IPY_MODEL_508fdbb60f5d419c93e07fc82add085a",
            "value": " 268M/268M [00:02&lt;00:00, 188MB/s]"
          }
        },
        "fee0aab961df4a6cb9cac50eee902401": {
          "model_module": "@jupyter-widgets/controls",
          "model_module_version": "1.5.0",
          "model_name": "DescriptionStyleModel",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}