{ "cells": [ { "cell_type": "code", "execution_count": 1, "id": "a8a62c24-f987-4eb0-889b-a1844f6e27b9", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: numpy in c:\\programdata\\anaconda3\\lib\\site-packages (2.3.5)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: pandas in c:\\programdata\\anaconda3\\lib\\site-packages (2.3.3)\n", "Requirement already satisfied: numpy>=1.26.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas) (2.3.5)\n", "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas) (2.9.0.post0)\n", "Requirement already satisfied: pytz>=2020.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas) (2025.2)\n", "Requirement already satisfied: tzdata>=2022.7 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas) (2025.2)\n", "Requirement already satisfied: six>=1.5 in c:\\programdata\\anaconda3\\lib\\site-packages (from python-dateutil>=2.8.2->pandas) (1.17.0)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: scikit-learn in c:\\programdata\\anaconda3\\lib\\site-packages (1.7.2)\n", "Requirement already satisfied: numpy>=1.22.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from scikit-learn) (2.3.5)\n", "Requirement already satisfied: scipy>=1.8.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from scikit-learn) (1.16.3)\n", "Requirement already satisfied: joblib>=1.2.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from scikit-learn) (1.5.2)\n", "Requirement already satisfied: threadpoolctl>=3.1.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from scikit-learn) (3.5.0)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: nltk in c:\\programdata\\anaconda3\\lib\\site-packages (3.9.2)\n", "Requirement already satisfied: matplotlib in c:\\programdata\\anaconda3\\lib\\site-packages (3.10.6)\n", "Requirement already satisfied: click in c:\\programdata\\anaconda3\\lib\\site-packages (from nltk) (8.2.1)\n", "Requirement already satisfied: joblib in c:\\programdata\\anaconda3\\lib\\site-packages (from nltk) (1.5.2)\n", "Requirement already satisfied: regex>=2021.8.3 in c:\\programdata\\anaconda3\\lib\\site-packages (from nltk) (2025.9.1)\n", "Requirement already satisfied: tqdm in c:\\programdata\\anaconda3\\lib\\site-packages (from nltk) (4.67.1)\n", "Requirement already satisfied: contourpy>=1.0.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib) (1.3.3)\n", "Requirement already satisfied: cycler>=0.10 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib) (0.11.0)\n", "Requirement already satisfied: fonttools>=4.22.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib) (4.60.1)\n", "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib) (1.4.9)\n", "Requirement already satisfied: numpy>=1.23 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib) (2.3.5)\n", "Requirement already satisfied: packaging>=20.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib) (25.0)\n", "Requirement already satisfied: pillow>=8 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib) (12.0.0)\n", "Requirement already satisfied: pyparsing>=2.3.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib) (3.2.5)\n", "Requirement already satisfied: python-dateutil>=2.7 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib) (2.9.0.post0)\n", "Requirement already satisfied: six>=1.5 in c:\\programdata\\anaconda3\\lib\\site-packages (from python-dateutil>=2.7->matplotlib) (1.17.0)\n", "Requirement already satisfied: colorama in c:\\programdata\\anaconda3\\lib\\site-packages (from click->nltk) (0.4.6)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: seaborn in c:\\programdata\\anaconda3\\lib\\site-packages (0.13.2)\n", "Requirement already satisfied: numpy!=1.24.0,>=1.20 in c:\\programdata\\anaconda3\\lib\\site-packages (from seaborn) (2.3.5)\n", "Requirement already satisfied: pandas>=1.2 in c:\\programdata\\anaconda3\\lib\\site-packages (from seaborn) (2.3.3)\n", "Requirement already satisfied: matplotlib!=3.6.1,>=3.4 in c:\\programdata\\anaconda3\\lib\\site-packages (from seaborn) (3.10.6)\n", "Requirement already satisfied: contourpy>=1.0.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.3.3)\n", "Requirement already satisfied: cycler>=0.10 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (0.11.0)\n", "Requirement already satisfied: fonttools>=4.22.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (4.60.1)\n", "Requirement already satisfied: kiwisolver>=1.3.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (1.4.9)\n", "Requirement already satisfied: packaging>=20.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (25.0)\n", "Requirement already satisfied: pillow>=8 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (12.0.0)\n", "Requirement already satisfied: pyparsing>=2.3.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (3.2.5)\n", "Requirement already satisfied: python-dateutil>=2.7 in c:\\programdata\\anaconda3\\lib\\site-packages (from matplotlib!=3.6.1,>=3.4->seaborn) (2.9.0.post0)\n", "Requirement already satisfied: pytz>=2020.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas>=1.2->seaborn) (2025.2)\n", "Requirement already satisfied: tzdata>=2022.7 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas>=1.2->seaborn) (2025.2)\n", "Requirement already satisfied: six>=1.5 in c:\\programdata\\anaconda3\\lib\\site-packages (from python-dateutil>=2.7->matplotlib!=3.6.1,>=3.4->seaborn) (1.17.0)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: transformers in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (5.3.0)\n", "Requirement already satisfied: huggingface-hub<2.0,>=1.3.0 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from transformers) (1.7.1)\n", "Requirement already satisfied: numpy>=1.17 in c:\\programdata\\anaconda3\\lib\\site-packages (from transformers) (2.3.5)\n", "Requirement already satisfied: packaging>=20.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from transformers) (25.0)\n", "Requirement already satisfied: pyyaml>=5.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from transformers) (6.0.3)\n", "Requirement already satisfied: regex!=2019.12.17 in c:\\programdata\\anaconda3\\lib\\site-packages (from transformers) (2025.9.1)\n", "Requirement already satisfied: tokenizers<=0.23.0,>=0.22.0 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from transformers) (0.22.2)\n", "Requirement already satisfied: typer in c:\\programdata\\anaconda3\\lib\\site-packages (from transformers) (0.20.0)\n", "Requirement already satisfied: safetensors>=0.4.3 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from transformers) (0.7.0)\n", "Requirement already satisfied: tqdm>=4.27 in c:\\programdata\\anaconda3\\lib\\site-packages (from transformers) (4.67.1)\n", "Requirement already satisfied: filelock>=3.10.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface-hub<2.0,>=1.3.0->transformers) (3.20.0)\n", "Requirement already satisfied: fsspec>=2023.5.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface-hub<2.0,>=1.3.0->transformers) (2025.10.0)\n", "Requirement already satisfied: hf-xet<2.0.0,>=1.4.2 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from huggingface-hub<2.0,>=1.3.0->transformers) (1.4.2)\n", "Requirement already satisfied: httpx<1,>=0.23.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface-hub<2.0,>=1.3.0->transformers) (0.28.1)\n", "Requirement already satisfied: typing-extensions>=4.1.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface-hub<2.0,>=1.3.0->transformers) (4.15.0)\n", "Requirement already satisfied: anyio in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1,>=0.23.0->huggingface-hub<2.0,>=1.3.0->transformers) (4.10.0)\n", "Requirement already satisfied: certifi in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1,>=0.23.0->huggingface-hub<2.0,>=1.3.0->transformers) (2025.11.12)\n", "Requirement already satisfied: httpcore==1.* in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1,>=0.23.0->huggingface-hub<2.0,>=1.3.0->transformers) (1.0.9)\n", "Requirement already satisfied: idna in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1,>=0.23.0->huggingface-hub<2.0,>=1.3.0->transformers) (3.11)\n", "Requirement already satisfied: h11>=0.16 in c:\\programdata\\anaconda3\\lib\\site-packages (from httpcore==1.*->httpx<1,>=0.23.0->huggingface-hub<2.0,>=1.3.0->transformers) (0.16.0)\n", "Requirement already satisfied: colorama in c:\\programdata\\anaconda3\\lib\\site-packages (from tqdm>=4.27->transformers) (0.4.6)\n", "Requirement already satisfied: sniffio>=1.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from anyio->httpx<1,>=0.23.0->huggingface-hub<2.0,>=1.3.0->transformers) (1.3.0)\n", "Requirement already satisfied: click>=8.0.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from typer->transformers) (8.2.1)\n", "Requirement already satisfied: shellingham>=1.3.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from typer->transformers) (1.5.4)\n", "Requirement already satisfied: rich>=10.11.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from typer->transformers) (14.2.0)\n", "Requirement already satisfied: markdown-it-py>=2.2.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from rich>=10.11.0->typer->transformers) (2.2.0)\n", "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from rich>=10.11.0->typer->transformers) (2.19.2)\n", "Requirement already satisfied: mdurl~=0.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from markdown-it-py>=2.2.0->rich>=10.11.0->typer->transformers) (0.1.2)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: torch in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (2.10.0)\n", "Requirement already satisfied: filelock in c:\\programdata\\anaconda3\\lib\\site-packages (from torch) (3.20.0)\n", "Requirement already satisfied: typing-extensions>=4.10.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from torch) (4.15.0)\n", "Requirement already satisfied: sympy>=1.13.3 in c:\\programdata\\anaconda3\\lib\\site-packages (from torch) (1.14.0)\n", "Requirement already satisfied: networkx>=2.5.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from torch) (3.5)\n", "Requirement already satisfied: jinja2 in c:\\programdata\\anaconda3\\lib\\site-packages (from torch) (3.1.6)\n", "Requirement already satisfied: fsspec>=0.8.5 in c:\\programdata\\anaconda3\\lib\\site-packages (from torch) (2025.10.0)\n", "Requirement already satisfied: setuptools in c:\\programdata\\anaconda3\\lib\\site-packages (from torch) (80.9.0)\n", "Requirement already satisfied: mpmath<1.4,>=1.1.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from sympy>=1.13.3->torch) (1.3.0)\n", "Requirement already satisfied: MarkupSafe>=2.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from jinja2->torch) (3.0.2)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: shap in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (0.51.0)\n", "Requirement already satisfied: numpy>=2 in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (2.3.5)\n", "Requirement already satisfied: scipy in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (1.16.3)\n", "Requirement already satisfied: scikit-learn in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (1.7.2)\n", "Requirement already satisfied: pandas in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (2.3.3)\n", "Requirement already satisfied: tqdm>=4.27.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (4.67.1)\n", "Requirement already satisfied: packaging>20.9 in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (25.0)\n", "Requirement already satisfied: slicer==0.0.8 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from shap) (0.0.8)\n", "Requirement already satisfied: numba in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (0.62.1)\n", "Requirement already satisfied: llvmlite in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (0.45.1)\n", "Requirement already satisfied: cloudpickle in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (3.1.1)\n", "Requirement already satisfied: typing-extensions in c:\\programdata\\anaconda3\\lib\\site-packages (from shap) (4.15.0)\n", "Requirement already satisfied: colorama in c:\\programdata\\anaconda3\\lib\\site-packages (from tqdm>=4.27.0->shap) (0.4.6)\n", "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas->shap) (2.9.0.post0)\n", "Requirement already satisfied: pytz>=2020.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas->shap) (2025.2)\n", "Requirement already satisfied: tzdata>=2022.7 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas->shap) (2025.2)\n", "Requirement already satisfied: six>=1.5 in c:\\programdata\\anaconda3\\lib\\site-packages (from python-dateutil>=2.8.2->pandas->shap) (1.17.0)\n", "Requirement already satisfied: joblib>=1.2.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from scikit-learn->shap) (1.5.2)\n", "Requirement already satisfied: threadpoolctl>=3.1.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from scikit-learn->shap) (3.5.0)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: tqdm in c:\\programdata\\anaconda3\\lib\\site-packages (4.67.1)\n", "Requirement already satisfied: colorama in c:\\programdata\\anaconda3\\lib\\site-packages (from tqdm) (0.4.6)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: accelerate in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (1.13.0)\n", "Requirement already satisfied: numpy>=1.17 in c:\\programdata\\anaconda3\\lib\\site-packages (from accelerate) (2.3.5)\n", "Requirement already satisfied: packaging>=20.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from accelerate) (25.0)\n", "Requirement already satisfied: psutil in c:\\programdata\\anaconda3\\lib\\site-packages (from accelerate) (7.0.0)\n", "Requirement already satisfied: pyyaml in c:\\programdata\\anaconda3\\lib\\site-packages (from accelerate) (6.0.3)\n", "Requirement already satisfied: torch>=2.0.0 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from accelerate) (2.10.0)\n", "Requirement already satisfied: huggingface_hub>=0.21.0 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from accelerate) (1.7.1)\n", "Requirement already satisfied: safetensors>=0.4.3 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from accelerate) (0.7.0)\n", "Requirement already satisfied: filelock>=3.10.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface_hub>=0.21.0->accelerate) (3.20.0)\n", "Requirement already satisfied: fsspec>=2023.5.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface_hub>=0.21.0->accelerate) (2025.10.0)\n", "Requirement already satisfied: hf-xet<2.0.0,>=1.4.2 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from huggingface_hub>=0.21.0->accelerate) (1.4.2)\n", "Requirement already satisfied: httpx<1,>=0.23.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface_hub>=0.21.0->accelerate) (0.28.1)\n", "Requirement already satisfied: tqdm>=4.42.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface_hub>=0.21.0->accelerate) (4.67.1)\n", "Requirement already satisfied: typer in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface_hub>=0.21.0->accelerate) (0.20.0)\n", "Requirement already satisfied: typing-extensions>=4.1.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface_hub>=0.21.0->accelerate) (4.15.0)\n", "Requirement already satisfied: anyio in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1,>=0.23.0->huggingface_hub>=0.21.0->accelerate) (4.10.0)\n", "Requirement already satisfied: certifi in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1,>=0.23.0->huggingface_hub>=0.21.0->accelerate) (2025.11.12)\n", "Requirement already satisfied: httpcore==1.* in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1,>=0.23.0->huggingface_hub>=0.21.0->accelerate) (1.0.9)\n", "Requirement already satisfied: idna in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1,>=0.23.0->huggingface_hub>=0.21.0->accelerate) (3.11)\n", "Requirement already satisfied: h11>=0.16 in c:\\programdata\\anaconda3\\lib\\site-packages (from httpcore==1.*->httpx<1,>=0.23.0->huggingface_hub>=0.21.0->accelerate) (0.16.0)\n", "Requirement already satisfied: sympy>=1.13.3 in c:\\programdata\\anaconda3\\lib\\site-packages (from torch>=2.0.0->accelerate) (1.14.0)\n", "Requirement already satisfied: networkx>=2.5.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from torch>=2.0.0->accelerate) (3.5)\n", "Requirement already satisfied: jinja2 in c:\\programdata\\anaconda3\\lib\\site-packages (from torch>=2.0.0->accelerate) (3.1.6)\n", "Requirement already satisfied: setuptools in c:\\programdata\\anaconda3\\lib\\site-packages (from torch>=2.0.0->accelerate) (80.9.0)\n", "Requirement already satisfied: mpmath<1.4,>=1.1.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from sympy>=1.13.3->torch>=2.0.0->accelerate) (1.3.0)\n", "Requirement already satisfied: colorama in c:\\programdata\\anaconda3\\lib\\site-packages (from tqdm>=4.42.1->huggingface_hub>=0.21.0->accelerate) (0.4.6)\n", "Requirement already satisfied: sniffio>=1.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from anyio->httpx<1,>=0.23.0->huggingface_hub>=0.21.0->accelerate) (1.3.0)\n", "Requirement already satisfied: MarkupSafe>=2.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from jinja2->torch>=2.0.0->accelerate) (3.0.2)\n", "Requirement already satisfied: click>=8.0.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from typer->huggingface_hub>=0.21.0->accelerate) (8.2.1)\n", "Requirement already satisfied: shellingham>=1.3.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from typer->huggingface_hub>=0.21.0->accelerate) (1.5.4)\n", "Requirement already satisfied: rich>=10.11.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from typer->huggingface_hub>=0.21.0->accelerate) (14.2.0)\n", "Requirement already satisfied: markdown-it-py>=2.2.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from rich>=10.11.0->typer->huggingface_hub>=0.21.0->accelerate) (2.2.0)\n", "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from rich>=10.11.0->typer->huggingface_hub>=0.21.0->accelerate) (2.19.2)\n", "Requirement already satisfied: mdurl~=0.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from markdown-it-py>=2.2.0->rich>=10.11.0->typer->huggingface_hub>=0.21.0->accelerate) (0.1.2)\n", "Defaulting to user installation because normal site-packages is not writeable\n", "Requirement already satisfied: datasets in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (4.7.0)\n", "Requirement already satisfied: filelock in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (3.20.0)\n", "Requirement already satisfied: numpy>=1.17 in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (2.3.5)\n", "Requirement already satisfied: pyarrow>=21.0.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (21.0.0)\n", "Requirement already satisfied: dill<0.4.1,>=0.3.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (0.4.0)\n", "Requirement already satisfied: pandas in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (2.3.3)\n", "Requirement already satisfied: requests>=2.32.2 in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (2.32.5)\n", "Requirement already satisfied: httpx<1.0.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (0.28.1)\n", "Requirement already satisfied: tqdm>=4.66.3 in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (4.67.1)\n", "Requirement already satisfied: xxhash in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from datasets) (3.6.0)\n", "Requirement already satisfied: multiprocess<0.70.19 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from datasets) (0.70.18)\n", "Requirement already satisfied: fsspec<=2026.2.0,>=2023.1.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from fsspec[http]<=2026.2.0,>=2023.1.0->datasets) (2025.10.0)\n", "Requirement already satisfied: huggingface-hub<2.0,>=0.25.0 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from datasets) (1.7.1)\n", "Requirement already satisfied: packaging in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (25.0)\n", "Requirement already satisfied: pyyaml>=5.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from datasets) (6.0.3)\n", "Requirement already satisfied: aiohttp!=4.0.0a0,!=4.0.0a1 in c:\\programdata\\anaconda3\\lib\\site-packages (from fsspec[http]<=2026.2.0,>=2023.1.0->datasets) (3.13.2)\n", "Requirement already satisfied: anyio in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1.0.0->datasets) (4.10.0)\n", "Requirement already satisfied: certifi in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1.0.0->datasets) (2025.11.12)\n", "Requirement already satisfied: httpcore==1.* in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1.0.0->datasets) (1.0.9)\n", "Requirement already satisfied: idna in c:\\programdata\\anaconda3\\lib\\site-packages (from httpx<1.0.0->datasets) (3.11)\n", "Requirement already satisfied: h11>=0.16 in c:\\programdata\\anaconda3\\lib\\site-packages (from httpcore==1.*->httpx<1.0.0->datasets) (0.16.0)\n", "Requirement already satisfied: hf-xet<2.0.0,>=1.4.2 in c:\\users\\user\\appdata\\roaming\\python\\python313\\site-packages (from huggingface-hub<2.0,>=0.25.0->datasets) (1.4.2)\n", "Requirement already satisfied: typer in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface-hub<2.0,>=0.25.0->datasets) (0.20.0)\n", "Requirement already satisfied: typing-extensions>=4.1.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from huggingface-hub<2.0,>=0.25.0->datasets) (4.15.0)\n", "Requirement already satisfied: aiohappyeyeballs>=2.5.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets) (2.6.1)\n", "Requirement already satisfied: aiosignal>=1.4.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets) (1.4.0)\n", "Requirement already satisfied: attrs>=17.3.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets) (25.4.0)\n", "Requirement already satisfied: frozenlist>=1.1.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets) (1.8.0)\n", "Requirement already satisfied: multidict<7.0,>=4.5 in c:\\programdata\\anaconda3\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets) (6.7.0)\n", "Requirement already satisfied: propcache>=0.2.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets) (0.3.1)\n", "Requirement already satisfied: yarl<2.0,>=1.17.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2026.2.0,>=2023.1.0->datasets) (1.22.0)\n", "Requirement already satisfied: charset_normalizer<4,>=2 in c:\\programdata\\anaconda3\\lib\\site-packages (from requests>=2.32.2->datasets) (3.4.4)\n", "Requirement already satisfied: urllib3<3,>=1.21.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from requests>=2.32.2->datasets) (2.5.0)\n", "Requirement already satisfied: colorama in c:\\programdata\\anaconda3\\lib\\site-packages (from tqdm>=4.66.3->datasets) (0.4.6)\n", "Requirement already satisfied: sniffio>=1.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from anyio->httpx<1.0.0->datasets) (1.3.0)\n", "Requirement already satisfied: python-dateutil>=2.8.2 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas->datasets) (2.9.0.post0)\n", "Requirement already satisfied: pytz>=2020.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas->datasets) (2025.2)\n", "Requirement already satisfied: tzdata>=2022.7 in c:\\programdata\\anaconda3\\lib\\site-packages (from pandas->datasets) (2025.2)\n", "Requirement already satisfied: six>=1.5 in c:\\programdata\\anaconda3\\lib\\site-packages (from python-dateutil>=2.8.2->pandas->datasets) (1.17.0)\n", "Requirement already satisfied: click>=8.0.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from typer->huggingface-hub<2.0,>=0.25.0->datasets) (8.2.1)\n", "Requirement already satisfied: shellingham>=1.3.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from typer->huggingface-hub<2.0,>=0.25.0->datasets) (1.5.4)\n", "Requirement already satisfied: rich>=10.11.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from typer->huggingface-hub<2.0,>=0.25.0->datasets) (14.2.0)\n", "Requirement already satisfied: markdown-it-py>=2.2.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from rich>=10.11.0->typer->huggingface-hub<2.0,>=0.25.0->datasets) (2.2.0)\n", "Requirement already satisfied: pygments<3.0.0,>=2.13.0 in c:\\programdata\\anaconda3\\lib\\site-packages (from rich>=10.11.0->typer->huggingface-hub<2.0,>=0.25.0->datasets) (2.19.2)\n", "Requirement already satisfied: mdurl~=0.1 in c:\\programdata\\anaconda3\\lib\\site-packages (from markdown-it-py>=2.2.0->rich>=10.11.0->typer->huggingface-hub<2.0,>=0.25.0->datasets) (0.1.2)\n" ] } ], "source": [ "#Install libraries\n", "!pip install numpy \n", "!pip install pandas \n", "!pip install scikit-learn \n", "!pip install nltk matplotlib \n", "!pip install seaborn \n", "!pip install transformers \n", "!pip install torch \n", "!pip install shap \n", "!pip install tqdm\n", "!pip install accelerate\n", "!pip install datasets" ] }, { "cell_type": "markdown", "id": "e1782168-17fb-486a-ba6f-1c89a9cc3f48", "metadata": {}, "source": [ "## Import libraries" ] }, { "cell_type": "code", "execution_count": 2, "id": "78f64821-b4cb-440d-b486-f3a9f6859176", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "[nltk_data] Downloading package stopwords to\n", "[nltk_data] C:\\Users\\User\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package stopwords is already up-to-date!\n", "[nltk_data] Downloading package punkt to\n", "[nltk_data] C:\\Users\\User\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package punkt is already up-to-date!\n", "[nltk_data] Downloading package wordnet to\n", "[nltk_data] C:\\Users\\User\\AppData\\Roaming\\nltk_data...\n", "[nltk_data] Package wordnet is already up-to-date!\n" ] } ], "source": [ "\n", "import os\n", "import numpy as np\n", "import pandas as pd\n", "from tqdm import tqdm\n", "\n", "import nltk\n", "nltk.download('stopwords')\n", "nltk.download('punkt')\n", "nltk.download('wordnet')\n", "\n", "import re\n", "from sklearn.feature_extraction.text import ENGLISH_STOP_WORDS\n", "\n", "from sklearn.model_selection import train_test_split\n", "from sklearn.feature_extraction.text import TfidfVectorizer\n", "from sklearn.linear_model import LogisticRegression\n", "from sklearn.metrics import accuracy_score, classification_report, confusion_matrix\n", "\n", "#BERT Model\n", "\n", "from transformers import (\n", " BertTokenizer,\n", " BertForSequenceClassification,\n", " Trainer\n", ")\n", "from transformers import TrainingArguments\n", "from collections import Counter\n", "import shap\n", "import torch\n", "import matplotlib.pyplot as plt\n", "import seaborn as sns\n", "from sklearn.svm import LinearSVC\n", "\n", "from sklearn.model_selection import GridSearchCV\n", "from sklearn.svm import LinearSVC" ] }, { "cell_type": "markdown", "id": "c09afa3c-c4ee-4e3d-b3d5-3ffb8776e324", "metadata": {}, "source": [ "## Load IMDB dataset" ] }, { "cell_type": "code", "execution_count": 3, "id": "40ac9538-4f2c-4f95-8bf9-dad45deb3aec", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "25000 25000\n" ] } ], "source": [ "\n", "DATASET_PATH = \"aclImdb\"\n", "\n", "def load_data(folder):\n", "\n", " texts = []\n", " labels = []\n", "\n", " for label in [\"pos\", \"neg\"]:\n", " path = os.path.join(folder, label)\n", "\n", " for file in os.listdir(path):\n", "\n", " with open(os.path.join(path, file), encoding=\"utf-8\") as f:\n", " texts.append(f.read())\n", "\n", " labels.append(1 if label == \"pos\" else 0)\n", "\n", " return texts, labels\n", "\n", "\n", "train_texts, train_labels = load_data(os.path.join(DATASET_PATH, \"train\"))\n", "test_texts, test_labels = load_data(os.path.join(DATASET_PATH, \"test\"))\n", "\n", "print(len(train_texts), len(test_texts))" ] }, { "cell_type": "markdown", "id": "a634964b-e6fc-45ac-a0fb-5d6cb88755d4", "metadata": {}, "source": [ "# Convert to DataFrame" ] }, { "cell_type": "code", "execution_count": 4, "id": "dcdca10b-b80c-4ddc-be66-02f015cf40fa", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
textlabel
0Bromwell High is a cartoon comedy. It ran at t...1
1Homelessness (or Houselessness as George Carli...1
2Brilliant over-acting by Lesley Ann Warren. Be...1
3This is easily the most underrated film inn th...1
4This is not the typical Mel Brooks film. It wa...1
\n", "
" ], "text/plain": [ " text label\n", "0 Bromwell High is a cartoon comedy. It ran at t... 1\n", "1 Homelessness (or Houselessness as George Carli... 1\n", "2 Brilliant over-acting by Lesley Ann Warren. Be... 1\n", "3 This is easily the most underrated film inn th... 1\n", "4 This is not the typical Mel Brooks film. It wa... 1" ] }, "execution_count": 4, "metadata": {}, "output_type": "execute_result" } ], "source": [ "\n", "train_df = pd.DataFrame({\n", " \"text\": train_texts,\n", " \"label\": train_labels\n", "})\n", "\n", "test_df = pd.DataFrame({\n", " \"text\": test_texts,\n", " \"label\": test_labels\n", "})\n", "\n", "train_df.head()\n" ] }, { "cell_type": "code", "execution_count": 5, "id": "7b4d9212-346f-445a-bc19-dbc6fa9404cc", "metadata": {}, "outputs": [], "source": [ "import matplotlib.pyplot as plt\n", "import seaborn as sns" ] }, { "cell_type": "markdown", "id": "316a7caa-6e2a-4a8d-9636-ba3a2268ee58", "metadata": {}, "source": [ "# EDA" ] }, { "cell_type": "code", "execution_count": 8, "id": "ee9c89b6-3da4-4854-9648-7844742fb0c8", "metadata": {}, "outputs": [ { "data": { "image/png": "iVBORw0KGgoAAAANSUhEUgAAAgsAAAGHCAYAAAA+xRHwAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjYsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvq6yFwwAAAAlwSFlzAAAPYQAAD2EBqD+naQAAScNJREFUeJzt3XlcVPX+x/H3YdgEAUFkX7VSU4RcchdJc9fUvFqZSeptccklLetmmFYubZZlda00H+aShWVplqa5pLmv5VIKiSa5oKAYCjPf3x/+mOs45zsyOOMMX97Px4PHvcwcZr7zguTDmXNmNCGEABEREZGEh6sXQERERO6NwwIRERHZxGGBiIiIbOKwQERERDZxWCAiIiKbOCwQERGRTRwWiIiIyCYOC0RERGQThwUiIiKyicMClcvcuXOhaZr5w9fXFxEREUhLS8OUKVNw6tQpq6+ZOHEiNE2z634uXbqEiRMn4qeffrLr6/TuKyEhAd26dbPrdm5kwYIFmDFjhu51mqZh4sSJDr0/R/vxxx/RuHFj+Pv7Q9M0fPXVV9Jtc3JyMHToUNxxxx2oUqUKQkJCkJSUhH//+9/Iyclx6jpnzZqFuXPnWl2enZ0NTdN0r3MnmzZtwsSJE3H+/Hm7vm7Dhg3o27cvoqOj4e3tjaCgILRo0QLvv/8+CgsLzdslJCQgPT3dsYsmuoanqxdAFducOXNQp04dFBcX49SpU9i4cSOmTZuG119/HYsXL0b79u3N2w4ZMgSdOnWy6/YvXbqEl156CQDQtm3bMn9dee6rPBYsWID9+/dj1KhRVtdt3rwZMTExTl9DeQkh0LdvX9xxxx1YtmwZ/P39Ubt2bd1tjx8/joYNG6JatWp4+umnUbt2beTn5+O3337D559/jqNHjyI2NtZpa501axZCQ0OtfiFGRkZi8+bNqFWrltPu2xE2bdqEl156Cenp6ahWrVqZviYjIwOTJk1CixYtMHnyZNSqVQuXLl0yDx6HDx/GW2+95dyFE/0/Dgt0U+rXr4/GjRubP7///vsxevRotGrVCr1798bvv/+O8PBwAEBMTIzTf3leunQJfn5+t+S+bqRZs2Yuvf8b+euvv5CXl4devXqhXbt2NredPXs2zpw5g61btyIxMdF8ec+ePfH888/DZDI5e7m6fHx83L5zeSxZsgSTJk3C4MGDMXv2bIu9ZJ07d8YzzzyDzZs3u3CFVOkIonKYM2eOACC2bdume/3nn38uAIiXXnrJfFlGRoa4/kfuxx9/FKmpqSIkJET4+vqK2NhY0bt3b1FYWCiysrIEAKuPgQMHWtzejh07xP333y+qVasmIiIipPcVHx8vunbtKjIzM0VSUpLw8fERiYmJ4u2339Z9bFlZWRaXr127VgAQa9euFUIIkZqaqru+UgBERkaGxW3s27dP9OjRQ1SrVk34+PiI5ORkMXfuXN37WbBggXj++edFZGSkCAgIEO3atRMHDx7U7X29DRs2iHvuuUdUrVpVVKlSRTRv3lx8++23Vt+Laz/i4+Oltzds2DDh4eEhLl68WKb737Ztm+jevbsIDg4WPj4+IiUlRSxevNhim9LOa9asEU888YSoXr26CAkJEb169RInTpwwbxcfHy9da+nPyJw5c6we2549e0SfPn1EYGCgCA4OFqNHjxbFxcXi4MGDomPHjqJq1aoiPj5eTJs2zWr9+fn54umnnxYJCQnCy8tLREVFiZEjR1o9fgBi2LBhYt68eaJOnTqiSpUqokGDBuKbb76x2franyM99evXF8HBwaKwsLBMvePj483/XQghxD///CPGjBkjkpOTzY+/WbNm4quvvrL62s8//1zcfffdIjAwUFSpUkUkJiaKRx991Hy90WgUkydPFnfccYfw9fUVQUFBIikpScyYMcPidg4fPiwefPBBUaNGDeHt7S3q1Kkj3n33XYttynpb5H64Z4GcokuXLjAYDFi/fr10m+zsbHTt2hWtW7fGJ598gmrVquHEiRNYuXIlrly5gsjISKxcuRKdOnXC4MGDMWTIEABAjRo1LG6nd+/eeOCBB/DEE09YPI+rZ/fu3Rg1ahQmTpyIiIgIfPbZZxg5ciSuXLmCsWPH2vUYZ82ahcceewxHjhzB0qVLb7j9oUOH0KJFC4SFheGdd95B9erVMX/+fKSnp+Pvv//GM888Y7H9888/j5YtW+Kjjz5CQUEBnn32WXTv3h0HDhyAwWCQ3s+6detw7733okGDBvj444/h4+ODWbNmoXv37li4cCH69euHIUOGIDk5Gb1798aIESPw0EMPwcfHR3qbzZs3x3vvvYfevXtjzJgxaN68OQIDA3W3Xbt2LTp16oSmTZvigw8+QFBQEBYtWoR+/frh0qVLVk8lDBkyBF27dsWCBQuQk5ODcePG4eGHH8aaNWsAAEuXLkWfPn0QFBSEWbNmAYDNtZbq27cvHn74YTz++ONYtWoVpk+fjuLiYqxevRpDhw7F2LFjsWDBAjz77LO47bbb0Lt3bwBX906lpqbi+PHjeP7559GgQQP8+uuvePHFF7Fv3z6sXr3a4i/95cuXY9u2bZg0aRKqVq2K6dOno1evXjh06BBq1qyJIUOGIC8vDzNnzkRmZiYiIyMBAHfeeafuuk+ePIn9+/ejX79+8PPzu+Hj1HP58mXk5eVh7NixiI6OxpUrV7B69Wr07t0bc+bMwSOPPALg6lNl/fr1Q79+/TBx4kT4+vrizz//NLcHgOnTp2PixIl44YUX0KZNGxQXF+PgwYMWx1/89ttvaNGiBeLi4vDGG28gIiIC33//PZ566imcOXMGGRkZZb4tclOunlaoYrrRngUhhAgPDxd169Y1f379X/tffPGFACB2794tvY3Tp0/r/oV+7e29+OKL0uuuFR8fLzRNs7q/e++9VwQGBpr/iivrngUhhOjatav0L/Lr1/3AAw8IHx8fcezYMYvtOnfuLPz8/MT58+ct7qdLly4W25Xurdm8ebPu/ZVq1qyZCAsLExcuXDBfVlJSIurXry9iYmKEyWQSQvzvr/LXXnvN5u0JIYTJZBKPP/648PDwEACEpmmibt26YvTo0Vad6tSpI+666y5RXFxscXm3bt1EZGSkMBqNQoj/dR46dKjFdtOnTxcAxMmTJ82X1atXT6Smplqty9aehTfeeMNi25SUFAFAZGZmmi8rLi4WNWrUEL179zZfNmXKFOHh4WH1s13687pixQrzZQBEeHi4KCgoMF+Wm5srPDw8xJQpU8yXvfbaa7o/U3p++eUXAUCMHz/+htuWun7PwvVKSkpEcXGxGDx4sLjrrrvMl7/++usCgPlnT0+3bt1ESkqKzfvv2LGjiImJEfn5+RaXDx8+XPj6+oq8vLwy3xa5J54NQU4jhLB5fUpKCry9vfHYY4/h008/xdGjR8t1P/fff3+Zt61Xrx6Sk5MtLnvooYdQUFCAnTt3luv+y2rNmjVo166d1YGA6enpuHTpktVz0D169LD4vEGDBgCAP//8U3ofhYWF2LJlC/r06YOqVauaLzcYDBgwYACOHz+OQ4cO2b12TdPwwQcf4OjRo5g1axYeffRRFBcX46233kK9evWwbt06AMAff/yBgwcPon///gCAkpIS80eXLl1w8uRJq/svz+Msi+vPfKlbty40TUPnzp3Nl3l6euK2226zuK9vv/0W9evXR0pKisX6O3bsCE3TrM7MSUtLQ0BAgPnz8PBwhIWF3fT6b9aSJUvQsmVLVK1aFZ6envDy8sLHH3+MAwcOmLdp0qQJgKt7YT7//HOcOHHC6nbuvvtu7NmzB0OHDsX333+PgoICi+uLiorw448/olevXvDz87P6nhcVFeGXX34p022R++KwQE5RWFiIs2fPIioqSrpNrVq1sHr1aoSFhWHYsGGoVasWatWqhbffftuu+yrdrVsWERER0svOnj1r1/3a6+zZs7prLW10/f1Xr17d4vPSXe///POP9D7OnTsHIYRd92OP+Ph4PPnkk/j444/x+++/Y/HixSgqKsK4ceMAAH///TcAYOzYsfDy8rL4GDp0KADgzJkzFrdZnsdZFiEhIRafe3t7w8/PD76+vlaXFxUVmT//+++/sXfvXqv1BwQEQAhxw/WXPobyrj8uLg4AkJWVVa6vB4DMzEzzKZfz58/H5s2bsW3bNgwaNMjisbZp0wZfffUVSkpK8MgjjyAmJgb169fHwoULzds899xzeP311/HLL7+gc+fOqF69Otq1a4ft27cDuPrzVFJSgpkzZ1o169KlC4D/fc9vdFvkvnjMAjnF8uXLYTQab3i6Y+vWrdG6dWsYjUZs374dM2fOxKhRoxAeHo4HHnigTPdlz2s35ObmSi8r/Ue/9JfJ5cuXLba7/peEvapXr46TJ09aXf7XX38BAEJDQ2/q9gEgODgYHh4eTr+fUn379sWUKVOwf/9+i9t+7rnnzMcAXE92eqa7CA0NRZUqVfDJJ59Ir3emyMhIJCUl4YcffjCf3WOv+fPnIzExEYsXL7b47+P6n2kAuO+++3Dffffh8uXL+OWXXzBlyhQ89NBDSEhIQPPmzeHp6YkxY8ZgzJgxOH/+PFavXo3nn38eHTt2RE5ODoKDg817roYNG6a7ntIzaG50W+U9RoOcj3sWyOGOHTuGsWPHIigoCI8//niZvsZgMKBp06Z47733AMD8lICj/sos9euvv2LPnj0Wly1YsAABAQFo2LAhgKsvcAMAe/futdhu2bJlVrdnz1+Q7dq1w5o1a8y/tEvNmzcPfn5+DjkF0N/fH02bNkVmZqbFukwmE+bPn4+YmBjccccddt+u3vABABcvXkROTo55r0Xt2rVx++23Y8+ePWjcuLHux7W77MvqZv5St1e3bt1w5MgRVK9eXXf9pT8f9rD353jChAk4d+4cnnrqKd2n8y5evIgffvhB+vWapsHb29tiUMjNzcXXX39tc42pqamYNm0aAGDXrl1W21SrVg19+vTBsGHDkJeXh+zsbPj5+SEtLQ27du1CgwYNdJvp7X3Ruy1yX9yzQDdl//795ucnT506hQ0bNmDOnDkwGAxYunSp1ZkL1/rggw+wZs0adO3aFXFxcSgqKjL/NVf6Yk4BAQGIj4/H119/jXbt2iEkJAShoaHl+gcbuLorvkePHpg4cSIiIyMxf/58rFq1CtOmTTP/VdOkSRPUrl0bY8eORUlJCYKDg7F06VJs3LjR6vaSkpKQmZmJ999/H40aNYKHh4fF605cKyMjA99++y3S0tLw4osvIiQkBJ999hmWL1+O6dOnIygoqFyP6XpTpkzBvffei7S0NIwdOxbe3t6YNWsW9u/fj4ULF9r9KpoA8Morr+Dnn39Gv379kJKSgipVqiArKwvvvvsuzp49i9dee8287YcffojOnTujY8eOSE9PR3R0NPLy8nDgwAHs3LkTS5Yssfv+k5KSsGjRIixevBg1a9aEr68vkpKS7L6dshg1ahS+/PJLtGnTBqNHj0aDBg1gMplw7Ngx/PDDD3j66afRtGlTu9cPAG+//TYGDhwILy8v1K5dWzo4/etf/8KECRMwefJkHDx4EIMHDza/KNOWLVvw4Ycfol+/fujQoYPu13fr1g2ZmZkYOnQo+vTpg5ycHEyePBmRkZH4/fffzdu9+OKLOH78ONq1a4eYmBicP38eb7/9Nry8vJCamgoA6N69u/n1VGrUqIE///wTM2bMQHx8PG6//Xbz42rVqhVat26NJ598EgkJCbhw4QL++OMPfPPNN+azK8pyW+SmXHt8JVVUpUeyl354e3uLsLAwkZqaKl599VVx6tQpq6+5/gyFzZs3i169eon4+Hjh4+MjqlevLlJTU8WyZcssvm716tXirrvuEj4+Prqvs3D69Okb3pcQ/3udhS+++ELUq1dPeHt7i4SEBPHmm29aff3hw4dFhw4dRGBgoKhRo4YYMWKEWL58udXZEHl5eaJPnz6iWrVqQtO0Mr3OQvfu3UVQUJDw9vYWycnJFkfyC/G/syGWLFlicbnekf8ypa+z4O/vL6pUqSKaNWtmce7/tbdXlrMhfvnlFzFs2DCRnJwsQkJChMFgEDVq1BCdOnWyODug1J49e0Tfvn1FWFiY8PLyEhEREeKee+4RH3zwgXkb2Rk1emedZGdniw4dOoiAgIAyv87C9T8XAwcOFP7+/lZrTU1NFfXq1bO47OLFi+KFF14QtWvXFt7e3ubXAxg9erTIzc01b4f/f52F6+mdnfDcc8+JqKgo8xkltl5nodS6detEnz59RGRkpPDy8hKBgYGiefPm4rXXXrM4A0Pv/qZOnSoSEhKEj4+PqFu3rpg9e7bVfxfffvut6Ny5s4iOjjb/N9ylSxexYcMG8zZvvPGGaNGihQgNDRXe3t4iLi5ODB48WGRnZ1vcX1ZWlhg0aJCIjo4WXl5eokaNGqJFixbi5Zdftvu2yP1oQtzgkHUiIiKq1HjMAhEREdnEYYGIiIhs4rBARERENnFYICIiIps4LBAREZFNHBaIiIjIpgr9okwmkwl//fUXAgICyvVCM0RERJWVEAIXLlxAVFQUPDxs7zuo0MPCX3/9ZfUOfkRERFR2OTk5iImJsblNhR4WSl8qNTs7G8HBwS5ejXsxGo349ddfUa9ePRgMBlcvx62wjRzbyLGNPnaRc/c2BQUFiI2NLdP7tVToV3AsKChAUFAQzp8/77DX1VdF6e4lPkVjjW3k2EaObfSxi5y7tyn9HZqfn4/AwECb21boPQul3PGb4Gqapt3wm19ZsY0c28ixjT52kVOpjRJnQxiNRlcvwe0YjUbs27ePbXSwjRzbyLGNPnaRU6mNEsMC6VPhB9RZ2EaObeTYRh+7yKnShsMCERER2cRhgYiIiGzi2RCKEkKgqKgIvr6+PAD0OmwjxzZybKOPXeTcvY09Z0Nwz4LCvL29Xb0Et8U2cmwjxzb62EVOlTZKDAsmk8nVS3A7JpMJ+/btYxsdbCPHNnJso49d5FRqo8SwQERERM7DYYGIiIhs4rBARERENvFsCEUJIWAymeDh4eGWR+G6EtvIsY0c2+hjFzl3b1Pp3hti6tSp8PX1dfUy3I6npydKSkpcvQy3xDZybCPHNvrYRc5RbTIyMhywmvJT4mkId5zYXE3TNERERLCNDraRYxs5ttHHLnIqtVFiWCAiIiLn4bBARERENnFYUFgFPnbV6dhGjm3k2EYfu8ip0kaJAxxV+WY4khACJ06ccPUy3BLbyLGNHNvoYxc5ldpwz4LCeIaIHNvIsY0c2+hjFzlV2igxLKhwpKmjaZqG0NBQttHBNnJsI8c2+thFTqU2SgwLRERE5DwcFoiIiMgmDgsK4yuqybGNHNvIsY0+dpFTpQ3PhlCUEAK5ubmuXoZbYhs5tpFjG33sIqdSG+5ZUJi/v7+rl+C22EaObeTYRh+7yKnSRolhQYUjTR1N0zQEBwezjQ62kWMbObbRxy5yKrVRYlggIiIi5+GwQERERDZxWFBYUVGRq5fgtthGjm3k2EYfu8ip0oZnQyhKCIEzZ864ehluiW3k2EaObfSxi5xKbbhnQWGBgYGuXoLbYhs5tpFjG33sIqdKGyWGBRWONHU0TdMQGBjINjrYRo5t5NhGH7vIqdRGiWGBiIiInIfDAhEREdmkxLDAAxytCSFQWFjINjrYRo5t5NhGH7vIqdRGiWGB9J07d87VS3BbbCPHNnJso49d5FRpw2FBYcHBwa5egttiGzm2kWMbfewip0obJYYFFY40dTRN0+Dv7882OthGjm3k2EYfu8ip1EaJYYGIiIich8MCERER2aTEsKDCkaaOJoRAQUEB2+hgGzm2kWMbfewip1IbJYYF0ldQUODqJbgttpFjGzm20ccucqq0UWJYUOHgEUfTNA2hoaFso4Nt5NhGjm30sYucSm2UGBZIn6+vr6uX4LbYRo5t5NhGH7vIqdKGwwIRERHZxGGBiIiIbFJiWFDhSFNHE0Lg3LlzbKODbeTYRo5t9LGLnEptlBgWSF9hYaGrl+C22EaObeTYRh+7yKnSRolhQYUjTR1N0zRERESwjQ62kWMbObbRxy5yKrVRYlggfZ6enq5egttiGzm2kWMbfewip0obDgtERERkE4cFIiIiskmJYUGFI00dTQiBM2fOsI0OtpFjGzm20ccuciq1UWJYIH1FRUWuXoLbYhs5tpFjG33sIqdKGyWGBRWONHU0TdMQHR3NNjrYRo5t5NhGH7vIqdRGiWGB9KnwA+osbCPHNnJso49d5FRpw2GBiIiIbOKwQERERDYpMSyocKSpowkhkJubyzY62EaObeTYRh+7yKnURolhgfQZjUZXL8FtsY0c28ixjT52kVOljRLDgioHkDiSSkfhOhrbyLGNHNvoYxc5ldooMSwQERGR83BYICIiIps4LBAREZFNSgwLKhxp6mhCCJw4cYJtdLCNHNvIsY0+dpFTqY0SwwLpMxgMrl6C22IbObaRYxt97CKnShslhgUVjjR1NE3TEBERwTY62EaObeTYRh+7yKnURolhgYiIiJyHwwIRERHZxGFBYSocVOMsbCPHNnJso49d5FRp4/JhYdasWUhMTISvry8aNWqEDRs22H0bqnwzHEmlo3AdjW3k2EaObfSxi5xKbVw6LCxevBijRo3Cf/7zH+zatQutW7dG586dcezYMVcuSxm+vr6uXoLbYhs5tpFjG33sIqdKG5cOC2+++SYGDx6MIUOGoG7dupgxYwZiY2Px/vvv23U7Khxp6miapiE0NJRtdLCNHNvIsY0+dpFTqY3LhoUrV65gx44d6NChg8XlHTp0wKZNm3S/5vLlyygoKLD4ICIiIudy2bBw5swZGI1GhIeHW1weHh6O3Nxc3a+ZMmUKgoKCzB+xsbG3YqlERESVmssPcLx+94wQQrrL5rnnnkN+fr75Iycn51YsscIqKSlx9RLcFtvIsY0c2+hjFzlV2ni66o5DQ0NhMBis9iKcOnXKam9DKR8fH/j4+FhdrsKRpo4mhJDuoans2EaObeTYRh+7yKnUxmV7Fry9vdGoUSOsWrXK4vJVq1ahRYsWLlqVWvz9/V29BLfFNnJsI8c2+thFTpU2Ln0aYsyYMfjoo4/wySef4MCBAxg9ejSOHTuGJ554wq7bUeFIU0fTNA3BwcFso4Nt5NhGjm30sYucSm1c9jQEAPTr1w9nz57FpEmTcPLkSdSvXx8rVqxAfHy8K5dFRERE13DpsAAAQ4cOxdChQ129DCIiIpJw+dkQ5DxFRUWuXoLbYhs5tpFjG33sIqdKG5fvWXAEng1hTQiBM2fOuHoZbolt5NhGjm30sYucSm24Z0FhgYGBrl6C22IbObaRYxt97CKnShslhgUVjjR1NE3TEBgYyDY62EaObeTYRh+7yKnURolhgYiIiJyHwwIRERHZpMSwwAMcrQkhUFhYyDY62EaObeTYRh+7yKnURolhgfSdO3fO1UtwW2wjxzZybKOPXeRUacNhQWHBwcGuXoLbYhs5tpFjG33sIqdKGyWGBRWONHU0TdPg7+/PNjrYRo5t5NhGH7vIqdRGiWGBiIiInIfDAhEREdmkxLCgwpGmjiaEQEFBAdvoYBs5tpFjG33sIqdSGyWGBdJXUFDg6iW4LbaRYxs5ttHHLnKqtFFiWFDh4BFH0zQNoaGhbKODbeTYRo5t9LGLnEptyvSuk8uWLSvzDfbo0aPciyHH8vX1dfUS3BbbyLGNHNvoYxc5VdqUaVjo2bNnmW5M0zQYjcabWQ8RERG5mTINCyaTydnrICIiIjd1U8csFBUVOWodN0WFI00dTQiBc+fOsY0OtpFjGzm20ccuciq1sXtYMBqNmDx5MqKjo1G1alUcPXoUADBhwgR8/PHHDl8glV9hYaGrl+C22EaObeTYRh+7yKnSxu5h4ZVXXsHcuXMxffp0eHt7my9PSkrCRx995NDFlZUKR5o6mqZpiIiIYBsdbCPHNnJso49d5FRqY/ewMG/ePPz3v/9F//79YTAYzJc3aNAABw8edOji6OZ4epbpkJRKiW3k2EaObfSxi5wqbeweFk6cOIHbbrvN6nKTyYTi4mKHLIqIiIjch93DQr169bBhwwary5csWYK77rrLIYsiIiIi92H3/pGMjAwMGDAAJ06cgMlkQmZmJg4dOoR58+bh22+/dcYab0iFI00dTQiBM2fOsI0OtpFjGzm20ccuciq1sXvPQvfu3bF48WKsWLECmqbhxRdfxIEDB/DNN9/g3nvvdcYaqZzc5dRWd8Q2cmwjxzb62EVOlTblep2Fjh07Yt26dbh48SIuXbqEjRs3okOHDo5eW5mpcKSpo2mahujoaLbRwTZybCPHNvrYRU6lNuU+THP79u04cOAANE1D3bp10ahRI0euixxAhR9QZ2EbObaRYxt97CKnShu7h4Xjx4/jwQcfxM8//4xq1aoBAM6fP48WLVpg4cKFiI2NdfQaiYiIyIXsfhpi0KBBKC4uxoEDB5CXl4e8vDwcOHAAQggMHjzYGWskIiIiF7J7z8KGDRuwadMm1K5d23xZ7dq1MXPmTLRs2dKhiysrFY40dTQhBHJzc9lGB9vIsY0c2+hjFzmV2ti9ZyEuLk73xZdKSkoQHR3tkEWRY/DtwuXYRo5t5NhGH7vIqdLG7mFh+vTpGDFiBLZv326elrZv346RI0fi9ddfd/gCy0KVA0gcSaWjcB2NbeTYRo5t9LGLnEptyvQ0RHBwsMWDLSwsRNOmTc2veV1SUgJPT08MGjQIPXv2dMpCiYiIyDXKNCzMmDHDycsgIiIid1WmYWHgwIHOXgcRERG5qZt678x//vnH6mDHwMDAm1pQeahwpKmjCSFw4sQJttHBNnJsI8c2+thFTqU2dh/gWFhYiOHDhyMsLAxVq1ZFcHCwxQe5D4PB4OoluC22kWMbObbRxy5yqrSxe1h45plnsGbNGsyaNQs+Pj746KOP8NJLLyEqKgrz5s1zxhpvSIUjTR1N0zRERESwjQ62kWMbObbRxy5yKrWx+2mIb775BvPmzUPbtm0xaNAgtG7dGrfddhvi4+Px2WefoX///s5YJxEREbmI3XsW8vLykJiYCODq8Ql5eXkAgFatWmH9+vWOXR0RERG5nN3DQs2aNZGdnQ0AuPPOO/H5558DuLrHofSNpcg9qHBQjbOwjRzbyLGNPnaRU6WN3cPCo48+ij179gAAnnvuOfOxC6NHj8a4ceMcvsCyUOWb4UgqHYXraGwjxzZybKOPXeRUamP3MQujR482//+0tDQcPHgQ27dvR61atZCcnOzQxdHN8fX1RVFRkauX4ZbYRo5t5NhGH7vIqdLG7j0L14uLi0Pv3r0REhKCQYMGOWJNdlPhSFNH0zQNoaGhbKODbeTYRo5t9LGLnEptbnpYKJWXl4dPP/3UUTdHREREbsJhwwIRERGpicOCwkpKSly9BLfFNnJsI8c2+thFTpU2N/XeEO5ChSNNHU0IgdzcXFcvwy2xjRzbyLGNPnaRU6lNmYeF3r1727z+/PnzN7sWcjB/f38UFha6ehluiW3k2EaObfSxi5wqbco8LAQFBd3w+kceeeSmF1QeKhxp6miapiE4OBiXLl3inpfrsI0c28ixjT52kVOpTZmHhTlz5jhzHUREROSmeIAjERER2cRhQWEqvGqYs7CNHNvIsY0+dpFTpY0mKvATKQUFBQgKCkJ+fj4CAwNdvRwiIqIKw57foUrsWTCZTK5egtsxmUzIzc1lGx1sI8c2cmyjj13kVGpj97Cwfv163ReZKCkpwfr16x2yKHtV4J0jTlN6fi/bWGMbObaRYxt97CKnUhu7h4W0tDTk5eVZXZ6fn4+0tDSHLIqIiIjch93DghBC93UNzp49C39/f4csioiIiNyH3a/gqGka0tPT4ePjY77OaDRi7969aNGiheNXWAZ8USZrmqYhJCSEbXSwjRzbyLGNPnaRU6mN3a/gKIRAQEAAqlSpYr7O29sbzZo1w7///W/Hr7AMPDyUOE7ToTw8PBAXF+fqZbgltpFjGzm20ccuciq1sfsVHBMSEjB27Fi3espBhSNNHc1kMuH48eOIiYnhMHUdtpFjGzm20ccuciq1sXv1GRkZbjUoADwbQo8QAnl5eWyjg23k2EaObfSxi5xKbeweFv7++28MGDAAUVFR8PT0hMFgsPggIiIitZT5aYhS6enpOHbsGCZMmIDIyEglDtwgIiIiObuHhY0bN2LDhg1ISUlxwnLKhwOLNU3TEBERwTY62EaObeTYRh+7yKnUxu5hITY21u2ef6noB444g4eHByIiIly9DLfENnJsI8c2+thFTqU2dv+WnTFjBsaPH4/s7GwnLKd8jEajq5fgdoxGI44cOcI2OthGjm3k2EYfu8ip1MbuPQv9+vXDpUuXUKtWLfj5+cHLy8vier2XgibXuHDhgquX4LbYRo5t5NhGH7vIqdLG7mFhxowZTlgGERERuSu7h4WBAwc6Yx1ERETkpsp1ZOCRI0fwwgsv4MEHH8SpU6cAACtXrsSvv/7q0MWVlQpHmjqapmmIjY1lGx1sI8c2cmyjj13kVGpj97Cwbt06JCUlYcuWLcjMzMTFixcBAHv37kVGRobDF1gWPBvCmoeHB6pXr842OthGjm3k2EYfu8ip1MbuRzB+/Hi8/PLLWLVqFby9vc2Xp6WlYfPmzQ5dXFmpcKSpoxmNRhw8eJBtdLCNHNvIsY0+dpFTqY3dw8K+ffvQq1cvq8tr1KiBs2fPOmRR5BhFRUWuXoLbYhs5tpFjG33sIqdKG7uHhWrVquHkyZNWl+/atQvR0dEOWRQRERG5D7uHhYceegjPPvsscnNzoWkaTCYTfv75Z4wdOxaPPPKIM9ZIRERELqQJO1+7ubi4GOnp6Vi0aBGEEPD09ITRaMRDDz2EuXPn3tJ3niwoKEBQUBDOnz+PoKCgW3a/FYEQAhcuXEBAQIASR+I6EtvIsY0c2+hjFzl3b1P6OzQ/Px+BgYE2t7V7WCh15MgR7Nq1CyaTCXfddRduv/32ci32ZtjzQImIiOh/7PkdWu7zOWrVqoU+ffqgb9++LhkUrqXCkaaOZjQasW/fPrbRwTZybCPHNvrYRU6lNna/gqMQAl988QXWrl2LU6dOwWQyWVyfmZnpsMXRzVHhB9RZ2EaObeTYRh+7yKnSxu5hYeTIkfjvf/+LtLQ0hIeHu+XzMEREROQ4dg8L8+fPR2ZmJrp06eKM9RAREZGbsfuYhaCgINSsWdMZayk3FV5K09E8PDxQu3ZtttHBNnJsI8c2+thFTqU2dj+CiRMn4qWXXsI///zjjPWQA137ctxkiW3k2EaObfSxi5wqbex+GuJf//oXFi5ciLCwMCQkJMDLy8vi+p07dzpscWWV3bAhgq470LKyM3l64kTv3ojOzIRHSYmrl+NW2EaObeTYRh+7yDmyTUJWloNWVT52Dwvp6enYsWMHHn74YR7gSEREVAnYPSwsX74c33//PVq1auWM9RAREZGbsfuYhdjYWL5aIhERUSVi98s9L1++HDNnzsQHH3yAhIQEJy2rbEpfqnJPXBwCFTja1JEEAOHpCa2kBHyiyBLbyLGNHNvoYxc5R7ZxxjEL9rzcs91PQzz88MO4dOkSatWqBT8/P6sDHPPy8uy9yZvH4yasaRqMfn7wvHABKN/bf6iLbeTYRo5t9LGLnEJt7B4WZsyY4YRl3BxhMAA8G8KCMBiQ26kTojMzofEIZQtsI8c2cmyjj13kVGpj97AwcOBAZ6yDiIiI3FSZhoWCggLz8xkFBQU2t+XBj0RERGop07AQHByMkydPIiwsDNWqVdN9bQUhBDRNU+YdtlTgUVzs6iW4LbaRYxs5ttHHLnKqtCnT2RDr1q1Dy5Yt4enpiXXr1tncNjU11WGLu5HSIzn3xsUhgGdDEBGRoirE2RDXDgCJiYmIjY212rsghEBOTk45lnvzBM+GsCI0DZfDw+Hz99/QKvhRuI7GNnJsI8c2+thFTqU2dv85npiYiNOnT1tdnpeXh8TERIcsyl7CYHDJ/bozYTDgdJs2bKODbeTYRo5t9LGLnEpt7B4WSo9NuN7Fixfh6+vrkEURERGR+yjzqZNjxowBAGiahgkTJsDPz898ndFoxJYtW5CSkuLwBRIREZFrlXlY2LVrF4Crexb27dtn8R7d3t7eSE5OxtixYx2/wrKo4M8FOYUQ8CooYBs9bCPHNnJso49d5BRqY/d7Qzz66KN4++233eL1FHg2BBERVQauPhvC7t+wc+bMcYtB4VqCg4IV4eGBi4mJbKODbeTYRo5t9LGLnEpt7H4EhYWFmDBhAlq0aIHbbrsNNWvWtPhwBRW+EY4mPDxwrkkTttHBNnJsI8c2+thFTqU2dr83xJAhQ7Bu3ToMGDAAkZGRumdGEBERkTrsHha+++47LF++HC1btnTGeoiIiMjN2L1vJDg4GCEhIc5YS/kpcKSpwwkB39xcttHDNnJsI8c2+thFTqE2dp8NMX/+fHz99df49NNPLV5rwRV4NgQREVUGrj4bwu6nId544w0cOXIE4eHhSEhIgJeXl8X1O3futPcmb5oKB484mvDwQEHdugg8cACayeTq5bgVtpFjGzm20ccuciq1sXtY6NmzpxOWcXOEhwdQwb8RjiY8PFBQrx4CDh2q8D+kjsY2cmwjxzb62EVOpTZ2DwsZGRnOWAcRERG5qXLtvz9//jw++ugjPPfcc8jLywNw9emHEydOOHRxRERE5Hp271nYu3cv2rdvj6CgIGRnZ+Pf//43QkJCsHTpUvz555+YN2+eM9ZpU0XfveMMmskE/6wsttHBNnJsI8c2+thFTqU2du9ZGDNmDNLT0/H7779bvCV1586dsX79eocurqxU+EY4mmYyIWTbNrbRwTZybCPHNvrYRU6lNnYPC9u2bcPjjz9udXl0dDRyc3Mdsih78WwIa8LDA3mKvMyoo7GNHNvIsY0+dpFTqY3dj8DX1xcFBQVWlx86dAg1atRwyKLspcI3wtGEhwcKFXkDE0djGzm2kWMbfewip1Ibux/Bfffdh0mTJqG4uBgAoGkajh07hvHjx+P+++93+AKJiIjIteweFl5//XWcPn0aYWFh+Oeff5CamorbbrsNAQEBeOWVV5yxRiIiInIhu8+GCAwMxMaNG7FmzRrs3LkTJpMJDRs2RPv27Z2xvjJR4eARR9NMJgT++ivb6GAbObaRYxt97CKnUhu73xvCnfC9IYiIqDJw9XtDlPk37JYtW/Ddd99ZXDZv3jwkJiYiLCwMjz32GC5fvly+Fd8kk8Hgkvt1ZyaDAafbtGEbHWwjxzZybKOPXeRUalPmYWHixInYu3ev+fN9+/Zh8ODBaN++PcaPH49vvvkGU6ZMccoib0jTXHO/7kzTUBQRwTZ62EaObeTYRh+7yCnUpszDwu7du9GuXTvz54sWLULTpk0xe/ZsjBkzBu+88w4+//xzpyySiIiIXKfMw8K5c+cQHh5u/nzdunXo1KmT+fMmTZogJyfHsasjIiIilyvzsBAeHo6s/z/A4sqVK9i5cyeaN29uvv7ChQvw8vJy/ArLQIUjTR1NM5kQrMjLjDoa28ixjRzb6GMXOZXalHlY6NSpE8aPH48NGzbgueeeg5+fH1q3bm2+fu/evahVq5ZTFnkjKnwjHE0zmVBVkTcwcTS2kWMbObbRxy5yKrUp87Dw8ssvw2AwIDU1FbNnz8bs2bPh7e1tvv6TTz5Bhw4dnLLIG1HhSFNHMxkMyO3UiW10sI0c28ixjT52kVOpTZlflKlGjRrYsGED8vPzUbVqVRiue/BLlixB1apVHb7AMtE0oOK+XIRzaBqKAwOVOArX4dhGjm3k2EYfu8gp1MbuV3AMCgrSvTwkJOSmF0NERETuhy97SERERDYpMSxoRqOrl+B2NKMRNdavZxsdbCPHNnJso49d5FRqY/fTEO5IE0KJ54QcSRMCvrm5rl6GW2IbObaRYxt97CKnUhsl9iyYPJWYeRzK5OmJE716sY0OtpFjGzm20ccuciq1UWJYIH0mF71IVkXANnJsI8c2+thFTpU2HBaIiIjIJg4LREREZJMmRMV9NaOCggIEBQVhT3w8AnmAowWhaSgJCIDnhQtXDwAlM7aRYxs5ttHHLnKObJPw/+/N5Eilv0Pz8/MRGBhoc1s19izwB9SaEDBcusQ2ethGjm3k2EYfu8gp1EaJYUEocKSpowlPT5zo3ZttdLCNHNvIsY0+dpFTqY0SwwIRERE5D4cFIiIisonDAhEREdmkxtkQcXEI9ODccy2Bq8+XaSUl4HkilthGjm3k2EYfu8g5sk2lPhti/fr16N69O6KioqBpGr766qvy3RBPm7SmaTD6+bGNHraRYxs5ttHHLnIKtXHpsFBYWIjk5GS8++67N3U7wmBw0IrUIQwG5HbqxDY62EaObeTYRh+7yKnUxqXnc3Tu3BmdO3d25RKIiIjoBirUyZ+XL1/G5cuXzZ8XFBS4cDVERESVQ4U6KnDKlCkICgoyf8TGxrp6SW7No7jY1UtwW2wjxzZybKOPXeRUaeM2Z0NomoalS5eiZ8+e0m309izExsZib1wcAng2BBERKapSnw1hLx8fHwQGBlp8AFffrIMsCU1DUUQE2+hgGzm2kWMbfewip1KbCjUsyKhwpKmjCYMBp9u0YRsdbCPHNnJso49d5FRq49IDHC9evIg//vjD/HlWVhZ2796NkJAQxMXFuXBlREREVMqlw8L27duRlpZm/nzMmDEAgIEDB2Lu3LkuWhURERFdy6XDQtu2beGQ4yvd4xhN9yIEvAoK2EYP28ixjRzb6GMXOYXauM3ZEOVReiQnz4YgIiKV8WwIBxAcFKwIDw9cTExkGx1sI8c2cmyjj13kVGpT8R8BOCzoER4eONekCdvoYBs5tpFjG33sIqdSm4r/CIiIiMipOCwQERGRTWoMCxX3GE3nEQK+ublso4dt5NhGjm30sYucQm14NgQREZGb49kQDqDCwSOOJjw8kF+vHtvoYBs5tpFjG33sIqdSm4r/CMBhQY/w8ECBIj+kjsY2cmwjxzb62EVOpTYV/xEQERGRU3FYICIiIpuUGBY0k8nVS3A7mskE/6wsttHBNnJsI8c2+thFTqU2PBuCiIjIzfFsCAdQ4eARRxMeHshT5GVGHY1t5NhGjm30sYucSm0q/iMAhwU9wsMDhYq8gYmjsY0c28ixjT52kVOpTcV/BERERORUHBaIiIjIJiWGBRWONHU0zWRC4K+/so0OtpFjGzm20ccuciq14dkQREREbo5nQziAyWBw9RLcjslgwOk2bdhGB9vIsY0c2+hjFzmV2igxLEDTXL0C96NpKIqIYBs9bCPHNnJso49d5BRqo8awQERERE7DYYGIiIhsUmJYUOFIU0fTTCYEb9vGNjrYRo5t5NhGH7vIqdSGZ0MQERG5OZ4N4QAqHGnqaCaDAbmdOrGNDraRYxs5ttHHLnIqtVFiWFDhSFOH0zQUBwayjR62kWMbObbRxy5yCrVRY1ggIiIip+GwQERERDYpMSxoRqOrl+B2NKMRNdavZxsdbCPHNnJso49d5FRq4+nqBTiCJoQSzwk5kiYEfHNzXb0Mt8Q2cmwjxzb62EVOpTZK7FkweSox8ziUydMTJ3r1YhsdbCPHNnJso49d5FRqo8SwQPpMXl6uXoLbYhs5tpFjG33sIqdKGw4LREREZBOHBSIiIrJJiZd73hMfj0Ae4GhBaBpKAgLgeeHC1QNAyYxt5NhGjm30sYucI9vw5Z4dgT+g1oSA4dIlttHDNnJsI8c2+thFTqE2SgwLQoEjTR1NeHriRO/ebKODbeTYRo5t9LGLnEptlBgWiIiIyHk4LBAREZFNHBaIiIjIJjXOhoiLQ6AH555rCVx9vkwrKQHPE7HENnJsI8c2+thFzpFteDaEI/C0SWuaBqOfH9voYRs5tpFjG33sIqdQGyWGBWEwuHoJbkcYDMjt1IltdLCNHNvIsY0+dpFTqY0SwwIRERE5D4cFIiIisonDgsI8iotdvQS3xTZybCPHNvrYRU6VNkqcDbE3Lg4BPBuCiIgUxbMhHEAocKSpowlNQ1FEBNvoYBs5tpFjG33sIqdSGzWGBQWONHU0YTDgdJs2bKODbeTYRo5t9LGLnEptlBgWiIiIyHk4LBAREZFNagwLFfcYTecRAl4FBWyjh23k2EaObfSxi5xCbXg2BBERkZtz9dkQng6/dxeI3bMH1apVc/Uy3IrJZMK5c+cQHBwMDw5SFthGjm3k2EYfu8ip1KZir/7/VeCdI04jhEBOTg7b6GAbObaRYxt97CKnUhslhgUiIiJyHg4LREREZBOHBYUFBAS4eglui23k2EaObfSxi5wqbZQ4G6IsR3ISERHR/1S694YwmUyuXoLbMZlMyM3NZRsdbCPHNnJso49d5FRqo8SwUIF3jjiNEAK5ublso4Nt5NhGjm30sYucSm2UGBaIiIjIeTgsEBERkU1KDAuaAu8V7miapiEkJIRtdLCNHNvIsY0+dpFTqQ3PhiAiIqqEeDYEwWQy4dixY2yjg23k2EaObfSxi5xKbZQYFirwzhGnEUIgLy+PbXSwjRzbyLGNPnaRU6mNEsMCEREROU+Ffovq0mmtoKAABoPBxatxL0ajERcvXmQbHWwjxzZybKOPXeTcvU1BQQGAsu2dr9DDwtmzZwEACQkJrl0IERFRBXXhwgUEBQXZ3KZCDwshISEAgGPHjt3wgVY2BQUFiI2NRU5ODs8UuQ7byLGNHNvoYxc5d28jhMCFCxcQFRV1w20r9LDg4XH1kIugoCC3/Ea4g8DAQLaRYBs5tpFjG33sIufObcr6hzYPcCQiIiKbOCwQERGRTRV6WPDx8UFGRgZ8fHxcvRS3wzZybCPHNnJso49d5FRqU6Ff7pmIiIicr0LvWSAiIiLn47BARERENnFYICIiIps4LBAREZFNFXpYmDVrFhITE+Hr64tGjRphw4YNrl6Sw0yZMgVNmjRBQEAAwsLC0LNnTxw6dMhiGyEEJk6ciKioKFSpUgVt27bFr7/+arHN5cuXMWLECISGhsLf3x89evTA8ePHLbY5d+4cBgwYgKCgIAQFBWHAgAE4f/68sx+iw0yZMgWapmHUqFHmyypzmxMnTuDhhx9G9erV4efnh5SUFOzYscN8fWVtU1JSghdeeAGJiYmoUqUKatasiUmTJlm8fXBlabN+/Xp0794dUVFR0DQNX331lcX1t7LDsWPH0L17d/j7+yM0NBRPPfUUrly54oyHXSa22hQXF+PZZ59FUlIS/P39ERUVhUceeQR//fWXxW0o2UZUUIsWLRJeXl5i9uzZ4rfffhMjR44U/v7+4s8//3T10hyiY8eOYs6cOWL//v1i9+7domvXriIuLk5cvHjRvM3UqVNFQECA+PLLL8W+fftEv379RGRkpCgoKDBv88QTT4jo6GixatUqsXPnTpGWliaSk5NFSUmJeZtOnTqJ+vXri02bNolNmzaJ+vXri27dut3Sx1teW7duFQkJCaJBgwZi5MiR5ssra5u8vDwRHx8v0tPTxZYtW0RWVpZYvXq1+OOPP8zbVNY2L7/8sqhevbr49ttvRVZWlliyZImoWrWqmDFjhnmbytJmxYoV4j//+Y/48ssvBQCxdOlSi+tvVYeSkhJRv359kZaWJnbu3ClWrVoloqKixPDhw53eQMZWm/Pnz4v27duLxYsXi4MHD4rNmzeLpk2bikaNGlnchoptKuywcPfdd4snnnjC4rI6deqI8ePHu2hFznXq1CkBQKxbt04IIYTJZBIRERFi6tSp5m2KiopEUFCQ+OCDD4QQV3+wvby8xKJFi8zbnDhxQnh4eIiVK1cKIYT47bffBADxyy+/mLfZvHmzACAOHjx4Kx5auV24cEHcfvvtYtWqVSI1NdU8LFTmNs8++6xo1aqV9PrK3KZr165i0KBBFpf17t1bPPzww0KIytvm+l+It7LDihUrhIeHhzhx4oR5m4ULFwofHx+Rn5/vlMdrD71B6npbt24VAMx/qKrapkI+DXHlyhXs2LEDHTp0sLi8Q4cO2LRpk4tW5Vz5+fkA/vfmWVlZWcjNzbVo4OPjg9TUVHODHTt2oLi42GKbqKgo1K9f37zN5s2bERQUhKZNm5q3adasGYKCgty+5bBhw9C1a1e0b9/e4vLK3GbZsmVo3Lgx/vWvfyEsLAx33XUXZs+ebb6+Mrdp1aoVfvzxRxw+fBgAsGfPHmzcuBFdunQBULnbXOtWdti8eTPq169v8UZGHTt2xOXLly2eOnNn+fn50DQN1apVA6Bumwr5RlJnzpyB0WhEeHi4xeXh4eHIzc110aqcRwiBMWPGoFWrVqhfvz4AmB+nXoM///zTvI23tzeCg4Ottin9+tzcXISFhVndZ1hYmFu3XLRoEXbu3Ilt27ZZXVeZ2xw9ehTvv/8+xowZg+effx5bt27FU089BR8fHzzyyCOVus2zzz6L/Px81KlTBwaDAUajEa+88goefPBBAJX75+Zat7JDbm6u1f0EBwfD29u7QrQqKirC+PHj8dBDD5nfKErVNhVyWCilaZrF50IIq8tUMHz4cOzduxcbN260uq48Da7fRm97d26Zk5ODkSNH4ocffoCvr690u8rYxmQyoXHjxnj11VcBAHfddRd+/fVXvP/++3jkkUfM21XGNosXL8b8+fOxYMEC1KtXD7t378aoUaMQFRWFgQMHmrerjG303KoOFbVVcXExHnjgAZhMJsyaNeuG21f0NhXyaYjQ0FAYDAar6erUqVNWk1hFN2LECCxbtgxr165FTEyM+fKIiAgAsNkgIiICV65cwblz52xu8/fff1vd7+nTp9225Y4dO3Dq1Ck0atQInp6e8PT0xLp16/DOO+/A09PTvO7K2CYyMhJ33nmnxWV169bFsWPHAFTun5tx48Zh/PjxeOCBB5CUlIQBAwZg9OjRmDJlCoDK3eZat7JDRESE1f2cO3cOxcXFbt2quLgYffv2RVZWFlatWmXx9tOqtqmQw4K3tzcaNWqEVatWWVy+atUqtGjRwkWrciwhBIYPH47MzEysWbMGiYmJFtcnJiYiIiLCosGVK1ewbt06c4NGjRrBy8vLYpuTJ09i//795m2aN2+O/Px8bN261bzNli1bkJ+f77Yt27Vrh3379mH37t3mj8aNG6N///7YvXs3atasWWnbtGzZ0uoU28OHDyM+Ph5A5f65uXTpEjw8LP/JMxgM5lMnK3Oba93KDs2bN8f+/ftx8uRJ8zY//PADfHx80KhRI6c+zvIqHRR+//13rF69GtWrV7e4Xtk2t/JoSkcqPXXy448/Fr/99psYNWqU8Pf3F9nZ2a5emkM8+eSTIigoSPz000/i5MmT5o9Lly6Zt5k6daoICgoSmZmZYt++feLBBx/UPb0pJiZGrF69WuzcuVPcc889uqfwNGjQQGzevFls3rxZJCUludVpXmVx7dkQQlTeNlu3bhWenp7ilVdeEb///rv47LPPhJ+fn5g/f755m8raZuDAgSI6Otp86mRmZqYIDQ0VzzzzjHmbytLmwoULYteuXWLXrl0CgHjzzTfFrl27zEf036oOpacHtmvXTuzcuVOsXr1axMTEuPTUSVttiouLRY8ePURMTIzYvXu3xb/Nly9fNt+Gim0q7LAghBDvvfeeiI+PF97e3qJhw4bm0wpVAED3Y86cOeZtTCaTyMjIEBEREcLHx0e0adNG7Nu3z+J2/vnnHzF8+HAREhIiqlSpIrp16yaOHTtmsc3Zs2dF//79RUBAgAgICBD9+/cX586duwWP0nGuHxYqc5tvvvlG1K9fX/j4+Ig6deqI//73vxbXV9Y2BQUFYuTIkSIuLk74+vqKmjVriv/85z8W/8hXljZr167V/fdl4MCBQohb2+HPP/8UXbt2FVWqVBEhISFi+PDhoqioyJkP3yZbbbKysqT/Nq9du9Z8Gyq24VtUExERkU0V8pgFIiIiunU4LBAREZFNHBaIiIjIJg4LREREZBOHBSIiIrKJwwIRERHZxGGBiIiIbOKwQERERDZxWCBSUHZ2NjRNw+7du129FLODBw+iWbNm8PX1RUpKikvXkpCQgBkzZrh0DUQVCYcFIidIT0+HpmmYOnWqxeVfffWV27/1rrNkZGTA398fhw4dwo8//qi7TWk3TdPg6emJuLg4PPnkk1bv4Heztm3bhscee8yht0mkMg4LRE7i6+uLadOmOfwXnStduXKl3F975MgRtGrVCvHx8Vbv1HetTp064eTJk8jOzsZHH32Eb775BkOHDi33/eqpUaMG/Pz8HHqbRCrjsEDkJO3bt0dERASmTJki3WbixIlWu+RnzJiBhIQE8+fp6eno2bMnXn31VYSHh6NatWp46aWXUFJSgnHjxiEkJAQxMTH45JNPrG7/4MGDaNGiBXx9fVGvXj389NNPFtf/9ttv6NKlC6pWrYrw8HAMGDAAZ86cMV/ftm1bDB8+HGPGjEFoaCjuvfde3cdhMpkwadIkxMTEwMfHBykpKVi5cqX5ek3TsGPHDkyaNAmapmHixInSJj4+PoiIiEBMTAw6dOiAfv364YcffrDYZs6cOahbty58fX1Rp04dzJo1y3xd8+bNMX78eIvtT58+DS8vL6xduxaA9dMQ+fn5eOyxxxAWFobAwEDcc8892LNnj/k6g8GAHTt2ALj69vEhISFo0qSJ+esXLlyIyMhIAFcHquHDhyMyMhK+vr5ISEiw+TNAVBFwWCByEoPBgFdffRUzZ87E8ePHb+q21qxZg7/++gvr16/Hm2++iYkTJ6Jbt24IDg7Gli1b8MQTT+CJJ55ATk6OxdeNGzcOTz/9NHbt2oUWLVqgR48eOHv2LADg5MmTSE1NRUpKCrZv346VK1fi77//Rt++fS1u49NPP4Wnpyd+/vlnfPjhh7rre/vtt/HGG2/g9ddfx969e9GxY0f06NEDv//+u/m+6tWrh6effhonT57E2LFjy/S4jx49ipUrV8LLy8t82ezZs/Gf//wHr7zyCg4cOIBXX30VEyZMwKeffgoA6N+/PxYuXIhr3yNv8eLFCA8PR2pqqtV9CCHQtWtX5ObmYsWKFdixYwcaNmyIdu3aIS8vD0FBQUhJSTEPWnv37jX/b0FBAQDgp59+Mt/2O++8g2XLluHzzz/HoUOHMH/+fIvhj6hCcsl7XRIpbuDAgeK+++4TQgjRrFkzMWjQICGEEEuXLhXX/meXkZEhkpOTLb72rbfeEvHx8Ra3FR8fL4xGo/my2rVri9atW5s/LykpEf7+/mLhwoVCCGF+K92pU6eatykuLhYxMTFi2rRpQgghJkyYIDp06GBx3zk5OQKAOHTokBDi6lt/p6Sk3PDxRkVFiVdeecXisiZNmoihQ4eaP09OThYZGRk2b2fgwIHCYDAIf39/4evra3773zfffNO8TWxsrFiwYIHF102ePFk0b95cCCHEqVOnhKenp1i/fr35+ubNm4tx48aZP4+PjxdvvfWWEEKIH3/8UQQGBlq99W+tWrXEhx9+KIQQYsyYMaJbt25CCCFmzJgh+vTpIxo2bCiWL18uhBDijjvuEO+//74QQogRI0aIe+65R5hMJpuPlagi4Z4FIiebNm0aPv30U/z222/lvo169erBw+N//7mGh4cjKSnJ/LnBYED16tVx6tQpi69r3ry5+f97enqicePGOHDgAABgx44dWLt2LapWrWr+qFOnDoCrxxeUaty4sc21FRQU4K+//kLLli0tLm/ZsqX5vuyRlpaG3bt3Y8uWLRgxYgQ6duyIESNGALj6dEJOTg4GDx5sse6XX37ZvOYaNWrg3nvvxWeffQYAyMrKwubNm9G/f3/d+9uxYwcuXryI6tWrW9xmVlaW+Tbbtm2LDRs2wGQyYd26dWjbti3atm2LdevWITc3F4cPHzbvWUhPT8fu3btRu3ZtPPXUU1ZPoRBVRJ6uXgCR6tq0aYOOHTvi+eefR3p6usV1Hh4eFrvLAaC4uNjqNq7dDQ9cPQZA7zKTyXTD9ZSejWEymdC9e3dMmzbNapvS598BwN/f/4a3ee3tlhJClOvMD39/f9x2220Aru7ST0tLw0svvYTJkyebH9/s2bPRtGlTi68zGAzm/9+/f3+MHDkSM2fOxIIFC1CvXj0kJyfr3p/JZEJkZKTV8RwAUK1aNQBXv4cXLlzAzp07sWHDBkyePBmxsbF49dVXkZKSgrCwMNStWxcA0LBhQ2RlZeG7777D6tWr0bdvX7Rv3x5ffPGF3S2I3AWHBaJbYOrUqUhJScEdd9xhcXmNGjWQm5tr8YvVka+N8Msvv6BNmzYAgJKSEuzYsQPDhw8HcPWX2pdffomEhAR4epb/n4LAwEBERUVh48aN5vsCgE2bNuHuu+++uQeAq6dcdu7cGU8++SSioqIQHR2No0ePSvcUAEDPnj3x+OOPY+XKlViwYAEGDBgg3bZhw4bIzc2Fp6en9NiC0uMW3n33XWiahjvvvBNRUVHYtWsXvv32W6tjIQIDA9GvXz/069cPffr0QadOnZCXl4eQkJByNSByNT4NQXQLJCUloX///pg5c6bF5W3btsXp06cxffp0HDlyBO+99x6+++47h93ve++9h6VLl+LgwYMYNmwYzp07h0GDBgEAhg0bhry8PDz44IPYunUrjh49ih9++AGDBg2C0Wi0637GjRuHadOmYfHixTh06BDGjx+P3bt3Y+TIkTf9GNq2bYt69erh1VdfBXD1DJIpU6bg7bffxuHDh7Fv3z7MmTMHb775pvlr/P39cd9992HChAk4cOAAHnroIentt2/fHs2bN0fPnj3x/fffIzs7G5s2bcILL7yA7du3W6xj/vz5SE1NhaZpCA4Oxp133onFixejbdu25u3eeustLFq0CAcPHsThw4exZMkSREREmPdSEFVEHBaIbpHJkydbPeVQt25dzJo1C++99x6Sk5OxdevWMp8pUBZTp07FtGnTkJycjA0bNuDrr79GaGgoACAqKgo///wzjEYjOnbsiPr162PkyJEICgqyOD6iLJ566ik8/fTTePrpp5GUlISVK1di2bJluP322x3yOMaMGYPZs2cjJycHQ4YMwUcffYS5c+ciKSkJqampmDt3LhITEy2+pn///tizZw9at26NuLg46W1rmoYVK1agTZs2GDRoEO644w488MADyM7ORnh4uHm7tLQ0GI1Gi8EgNTUVRqPRYs9C1apVMW3aNDRu3BhNmjRBdnY2VqxYYXdTIneiiev/9SIiIiK6BkddIiIisonDAhEREdnEYYGIiIhs4rBARERENnFYICIiIps4LBAREZFNHBaIiIjIJg4LREREZBOHBSIiIrKJwwIRERHZxGGBiIiIbPo/OqjKnynN5m4AAAAASUVORK5CYII=", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "plt.figure(figsize=(6,4))\n", "\n", "sns.countplot(\n", " y=train_df[\"label\"],\n", " hue=train_df[\"label\"], # fix warning\n", " palette=[\"Gray\", \"red\"],\n", " legend=False\n", ")\n", "\n", "plt.title(\"Distribution of Sentiment Classes\", fontsize=12)\n", "plt.xlabel(\"Number of Reviews\")\n", "plt.ylabel(\"Sentiment Label\")\n", "\n", "plt.grid(axis=\"x\", linestyle=\"--\", alpha=0.6)\n", "\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "7e7e7b8b-2a67-4c45-83fe-e168b37e29cf", "metadata": {}, "source": [ "# Preprocessing" ] }, { "cell_type": "code", "execution_count": 9, "id": "57293827-9a64-4930-8f68-4034edaaf495", "metadata": {}, "outputs": [ { "data": { "text/html": [ "
\n", "\n", "\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
textlabelclean
0Bromwell High is a cartoon comedy. It ran at t...1bromwell high cartoon comedy ran time programs...
1Homelessness (or Houselessness as George Carli...1homelessness houselessness george carlin state...
2Brilliant over-acting by Lesley Ann Warren. Be...1brilliant acting lesley ann warren best dramat...
3This is easily the most underrated film inn th...1easily underrated film inn brooks cannon sure ...
4This is not the typical Mel Brooks film. It wa...1typical mel brooks film slapstick movies actua...
\n", "
" ], "text/plain": [ " text label \\\n", "0 Bromwell High is a cartoon comedy. It ran at t... 1 \n", "1 Homelessness (or Houselessness as George Carli... 1 \n", "2 Brilliant over-acting by Lesley Ann Warren. Be... 1 \n", "3 This is easily the most underrated film inn th... 1 \n", "4 This is not the typical Mel Brooks film. It wa... 1 \n", "\n", " clean \n", "0 bromwell high cartoon comedy ran time programs... \n", "1 homelessness houselessness george carlin state... \n", "2 brilliant acting lesley ann warren best dramat... \n", "3 easily underrated film inn brooks cannon sure ... \n", "4 typical mel brooks film slapstick movies actua... " ] }, "execution_count": 9, "metadata": {}, "output_type": "execute_result" } ], "source": [ "\n", "\n", "stop_words = set(ENGLISH_STOP_WORDS)\n", "\n", "def clean_text(text):\n", "\n", " # lower case\n", " text = text.lower()\n", "\n", " # remove numbers + punctuation\n", " text = re.sub(r\"[^a-zA-Z]\", \" \", text)\n", "\n", " # split words\n", " words = text.split()\n", "\n", " # remove stopwords\n", " words = [w for w in words if w not in stop_words]\n", "\n", " # join back\n", " return \" \".join(words)\n", "\n", "train_df[\"clean\"] = train_df[\"text\"].apply(clean_text)\n", "test_df[\"clean\"] = test_df[\"text\"].apply(clean_text)\n", "\n", "train_df.head()" ] }, { "cell_type": "markdown", "id": "a448b143-26af-4be3-89aa-95ca1a6dcc39", "metadata": {}, "source": [ "# TF-IDF" ] }, { "cell_type": "code", "execution_count": 10, "id": "7e9b1b5b-ccb6-443a-9fc7-7c11e70792fb", "metadata": {}, "outputs": [], "source": [ "\n", "tfidf = TfidfVectorizer(\n", " max_features=10000,\n", " ngram_range=(1,2)\n", ")\n", "\n", "X_train = tfidf.fit_transform(train_df[\"clean\"])\n", "X_test = tfidf.transform(test_df[\"clean\"])\n", "\n", "y_train = train_df[\"label\"]\n", "y_test = test_df[\"label\"]" ] }, { "cell_type": "markdown", "id": "26eb6671-9891-4d37-94c6-7afc730f04c3", "metadata": {}, "source": [ "## SVM Model" ] }, { "cell_type": "code", "execution_count": 11, "id": "acc448e0-9318-4df7-83e0-3035ec544635", "metadata": {}, "outputs": [], "source": [ "model_svm = LinearSVC()\n", "\n", "model_svm.fit(X_train, y_train)\n", "\n", "pred_svm = model_svm.predict(X_test)" ] }, { "cell_type": "markdown", "id": "aea8fbdf-2610-4863-a86f-3036de2c7817", "metadata": {}, "source": [ "## Evaluation" ] }, { "cell_type": "code", "execution_count": 12, "id": "f3a011a1-b17b-45ca-8ad3-83e7ecb6ffab", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Accuracy: 0.86584\n", " precision recall f1-score support\n", "\n", " 0 0.86 0.88 0.87 12500\n", " 1 0.87 0.86 0.86 12500\n", "\n", " accuracy 0.87 25000\n", " macro avg 0.87 0.87 0.87 25000\n", "weighted avg 0.87 0.87 0.87 25000\n", "\n" ] } ], "source": [ "\n", "print(\"Accuracy:\", accuracy_score(y_test, pred_svm))\n", "\n", "print(classification_report(y_test, pred_svm))" ] }, { "cell_type": "markdown", "id": "140bf53b-4ada-4d02-abed-7198a3e05d9e", "metadata": {}, "source": [ "## Confusion matrix" ] }, { "cell_type": "code", "execution_count": 13, "id": "39ab1903-124a-4e6d-8831-347020cfb0ec", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "\n", "cm = confusion_matrix(y_test, pred_svm)\n", "\n", "plt.figure(figsize=(5,5))\n", "\n", "sns.heatmap(\n", " cm,\n", " annot=True,\n", " fmt=\"d\",\n", " cmap=\"coolwarm\", # changed colour\n", " linewidths=1, # added border\n", " linecolor=\"black\",\n", " cbar=True\n", ")\n", "\n", "plt.title(\"SVM Confusion Matrix Result\", fontsize=12)\n", "plt.xlabel(\"Predicted Label\", fontsize=10)\n", "plt.ylabel(\"True Label\", fontsize=10)\n", "\n", "plt.xticks([0.5,1.5], [\"Negative\", \"Positive\"])\n", "plt.yticks([0.5,1.5], [\"Negative\", \"Positive\"])\n", "\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "65bb3e33-95b8-4a3e-9171-a9acf1857236", "metadata": {}, "source": [ "## Hyperparameter tuning" ] }, { "cell_type": "code", "execution_count": 14, "id": "fe13650e-fe8c-40e7-abf2-206d0357ee83", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Best params: {'C': 0.1}\n", " precision recall f1-score support\n", "\n", " 0 0.89 0.88 0.88 12500\n", " 1 0.88 0.89 0.88 12500\n", "\n", " accuracy 0.88 25000\n", " macro avg 0.88 0.88 0.88 25000\n", "weighted avg 0.88 0.88 0.88 25000\n", "\n" ] } ], "source": [ "\n", "params = {\n", " \"C\": [0.01, 0.1, 1, 10]\n", "}\n", "\n", "grid = GridSearchCV(\n", " LinearSVC(),\n", " params,\n", " cv=3,\n", " n_jobs=-1\n", ")\n", "\n", "grid.fit(X_train, y_train)\n", "\n", "print(\"Best params:\", grid.best_params_)\n", "best_svm = grid.best_estimator_\n", "\n", "pred_svm = best_svm.predict(X_test)\n", "\n", "print(classification_report(y_test, pred_svm))" ] }, { "cell_type": "markdown", "id": "6e544989-5549-44a8-b2b1-3319af4bcd61", "metadata": {}, "source": [ "# Tokenizer" ] }, { "cell_type": "code", "execution_count": 15, "id": "e91775f3-ca0c-4668-9f44-0604c8b630b1", "metadata": {}, "outputs": [], "source": [ "\n", "tokenizer = BertTokenizer.from_pretrained(\n", " \"bert-base-uncased\"\n", ")" ] }, { "cell_type": "markdown", "id": "6341d9e9-eadb-4c03-8cae-393a130d6927", "metadata": {}, "source": [ "# Prepare small dataset" ] }, { "cell_type": "code", "execution_count": 16, "id": "b3e03305-5e8f-4d68-88d6-203c6c4ca0a6", "metadata": {}, "outputs": [], "source": [ "\n", "train_small = train_df.sample(4000)\n", "test_small = test_df.sample(1000)" ] }, { "cell_type": "markdown", "id": "40b8e5a7-3770-4145-b5ba-b3b8e8a48b3d", "metadata": {}, "source": [ "# Encode" ] }, { "cell_type": "code", "execution_count": 17, "id": "87ca2618-3db9-4688-b340-8304e1822ce4", "metadata": {}, "outputs": [], "source": [ "\n", "def tokenize(batch):\n", "\n", " return tokenizer(\n", " batch,\n", " padding=True,\n", " truncation=True,\n", " max_length=128\n", " )\n", "\n", "\n", "train_enc = tokenize(list(train_small[\"text\"]))\n", "test_enc = tokenize(list(test_small[\"text\"]))" ] }, { "cell_type": "markdown", "id": "5a928948-440a-4d1d-85e3-e8d93de73e7f", "metadata": {}, "source": [ "# Torch dataset" ] }, { "cell_type": "code", "execution_count": 18, "id": "f819632c-c525-40d2-818e-f55c8ff2c441", "metadata": {}, "outputs": [], "source": [ "\n", "class Dataset(torch.utils.data.Dataset):\n", "\n", " def __init__(self, encodings, labels):\n", "\n", " self.encodings = encodings\n", " self.labels = labels\n", "\n", " def __getitem__(self, idx):\n", "\n", " item = {\n", " key: torch.tensor(val[idx])\n", " for key, val in self.encodings.items()\n", " }\n", "\n", " item[\"labels\"] = torch.tensor(self.labels[idx])\n", "\n", " return item\n", "\n", " def __len__(self):\n", "\n", " return len(self.labels)\n", "\n", "\n", "train_dataset = Dataset(\n", " train_enc,\n", " list(train_small[\"label\"])\n", ")\n", "\n", "test_dataset = Dataset(\n", " test_enc,\n", " list(test_small[\"label\"])\n", ")" ] }, { "cell_type": "markdown", "id": "970343a2-86e8-46cf-bc94-8eb3645d6538", "metadata": {}, "source": [ "# Load BERT" ] }, { "cell_type": "code", "execution_count": 19, "id": "2c7dac00-f91e-45e9-8020-81707b3f7dee", "metadata": {}, "outputs": [ { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "d7f1c36738a5475e9551d6e8ef88f2be", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Loading weights: 0%| | 0/199 [00:00\n", " \n", " \n", " [500/500 44:18, Epoch 1/1]\n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", " \n", "
StepTraining Loss
500.666952
1000.418741
1500.429561
2000.444826
2500.471146
3000.363600
3500.461753
4000.340802
4500.334505
5000.409033

" ], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "data": { "application/vnd.jupyter.widget-view+json": { "model_id": "dfad5f0c3970428ca187efccf34a04f2", "version_major": 2, "version_minor": 0 }, "text/plain": [ "Writing model shards: 0%| | 0/1 [00:00" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "loss_values = []\n", "\n", "for log in trainer.state.log_history:\n", " if \"loss\" in log:\n", " loss_values.append(log[\"loss\"])\n", "\n", "plt.figure(figsize=(6,4))\n", "\n", "plt.plot(\n", " loss_values,\n", " marker=\"o\",\n", " color=\"darkred\"\n", ")\n", "\n", "plt.title(\"BERT Training Loss Curve\")\n", "plt.xlabel(\"Training Step\")\n", "plt.ylabel(\"Loss\")\n", "\n", "plt.grid(True, linestyle=\"--\", alpha=0.6)\n", "\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "dd596851-e8ca-4488-adb7-f19ac9ee2fe0", "metadata": {}, "source": [ "# Evaluate BERT" ] }, { "cell_type": "code", "execution_count": 26, "id": "ef10d416-6965-4a85-897f-63ab6ff641ca", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "C:\\Users\\User\\AppData\\Roaming\\Python\\Python313\\site-packages\\torch\\utils\\data\\dataloader.py:775: UserWarning: 'pin_memory' argument is set as true but no accelerator is found, then device pinned memory won't be used.\n", " super().__init__(loader)\n" ] }, { "data": { "text/html": [], "text/plain": [ "" ] }, "metadata": {}, "output_type": "display_data" }, { "name": "stdout", "output_type": "stream", "text": [ " precision recall f1-score support\n", "\n", " 0 0.86 0.85 0.86 487\n", " 1 0.86 0.87 0.86 513\n", "\n", " accuracy 0.86 1000\n", " macro avg 0.86 0.86 0.86 1000\n", "weighted avg 0.86 0.86 0.86 1000\n", "\n" ] } ], "source": [ "\n", "pred = trainer.predict(test_dataset)\n", "\n", "pred_labels = np.argmax(pred.predictions, axis=1)\n", "\n", "print(classification_report(\n", " test_small[\"label\"],\n", " pred_labels\n", "))" ] }, { "cell_type": "markdown", "id": "92f0d67f-6835-4e9a-88a4-f1790afc9cb5", "metadata": {}, "source": [ "# SHAP" ] }, { "cell_type": "code", "execution_count": 27, "id": "6a641928-c8fc-4265-aa9d-09ff20814f71", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "C:\\Users\\User\\AppData\\Local\\Temp\\ipykernel_8328\\2378116444.py:8: FutureWarning: The NumPy global RNG was seeded by calling `np.random.seed`. In a future version this function will no longer use the global RNG. Pass `rng` explicitly to opt-in to the new behaviour and silence this warning.\n", " shap.summary_plot(\n" ] }, { "data": { "image/png": "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", "text/plain": [ "

" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "\n", "# create explainer\n", "explainer = shap.Explainer(model_svm, X_train)\n", "\n", "# explain some samples\n", "shap_values = explainer(X_test[:100])\n", "\n", "# plot\n", "shap.summary_plot(\n", " shap_values,\n", " X_test[:100],\n", " plot_type=\"bar\",\n", " color=\"purple\"\n", ")" ] }, { "cell_type": "markdown", "id": "c68b637e-df48-48dd-80c2-9f7c81eb3714", "metadata": {}, "source": [ "# Word frequency" ] }, { "cell_type": "code", "execution_count": 28, "id": "a8d87123-5036-41f9-b048-af144199f88d", "metadata": {}, "outputs": [ { "data": { "text/plain": [ "[('br', 101872),\n", " ('s', 66099),\n", " ('movie', 44047),\n", " ('film', 40162),\n", " ('t', 34407),\n", " ('like', 20281),\n", " ('just', 17774),\n", " ('good', 15147),\n", " ('time', 12727),\n", " ('story', 11990),\n", " ('really', 11739),\n", " ('bad', 9308),\n", " ('people', 9287),\n", " ('great', 9061),\n", " ('don', 8848),\n", " ('way', 8026),\n", " ('make', 8025),\n", " ('movies', 7668),\n", " ('think', 7298),\n", " ('characters', 7159)]" ] }, "execution_count": 28, "metadata": {}, "output_type": "execute_result" } ], "source": [ "all_words = \" \".join(train_df[\"clean\"])\n", "\n", "words = all_words.split()\n", "\n", "Counter(words).most_common(20)" ] }, { "cell_type": "code", "execution_count": null, "id": "0f6a4bd1-9f74-4955-ba8a-7fceb274663e", "metadata": {}, "outputs": [], "source": [] } ], "metadata": { "kernelspec": { "display_name": "Python 3 (ipykernel)", "language": "python", "name": "python3" }, "language_info": { "codemirror_mode": { "name": "ipython", "version": 3 }, "file_extension": ".py", "mimetype": "text/x-python", "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", "version": "3.13.9" } }, "nbformat": 4, "nbformat_minor": 5 }