explorative_analysis.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2024-11-28T21:26:46.650474Z",
     "start_time": "2024-11-28T21:26:38.265802Z"
    }
   },
   "source": [
    "# Importing necessary libraries\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import pathlib\n",
    "\n",
    "from dotenv import load_dotenv\n",
    "from transformers import AutoTokenizer  # Correct import for Hugging Face tokenizer\n",
    "\n",
    "from src.dataloader import DataManager\n",
    "\n",
    "\n",
    "# Define the static method to read misconceptions from a file\n",
    "class MisconceptionAnalyzer:\n",
    "    @staticmethod\n",
    "    def _get_misconceptions(file_path: pathlib.Path) -> pd.DataFrame:\n",
    "        misconceptions = pd.read_csv(file_path)\n",
    "        # Ensuring columns match the actual file structure\n",
    "        misconceptions.columns = ['misconception_id', 'misconception']\n",
    "        return misconceptions\n",
    "\n",
    "# File path to the misconceptions file\n",
    "file_path = pathlib.Path('data/misconception_mapping.csv')\n",
    "\n",
    "# Load the misconceptions data\n",
    "misconceptions_df = MisconceptionAnalyzer._get_misconceptions(file_path)\n",
    "\n",
    "# Add a new column for word length of misconceptions\n",
    "misconceptions_df['word_length'] = misconceptions_df['misconception'].apply(lambda x: len(x.split()))\n",
    "\n",
    "# Optional: Add a new column for tokenized length of misconceptions\n",
    "# Initialize the tokenizer using Hugging Face transformers\n",
    "tokenizer = AutoTokenizer.from_pretrained(\"bert-base-uncased\")  # Example tokenizer\n",
    "\n",
    "misconceptions_df['token_length'] = misconceptions_df['misconception'].apply(\n",
    "    lambda x: len(tokenizer.tokenize(x))\n",
    ")\n",
    "\n",
    "# Sort the DataFrame by word length\n",
    "misconceptions_df = misconceptions_df.sort_values(by='word_length')\n",
    "\n",
    "# Plotting the word length of misconceptions\n",
    "plt.figure(figsize=(12, 8))\n",
    "plt.bar(\n",
    "    misconceptions_df['misconception_id'].astype(str),\n",
    "    misconceptions_df['word_length'],\n",
    "    color='blue',\n",
    "    label='Word Length'\n",
    ")\n",
    "\n",
    "# Optional: Add tokenized lengths to the plot for comparison\n",
    "plt.bar(\n",
    "    misconceptions_df['misconception_id'].astype(str),\n",
    "    misconceptions_df['token_length'],\n",
    "    color='orange',\n",
    "    alpha=0.6,\n",
    "    label='Tokenized Length'\n",
    ")\n",
    "\n",
    "plt.xlabel('Misconception ID (sorted by Word Length)')\n",
    "plt.ylabel('Length')\n",
    "plt.title('Word and Tokenized Length of Misconceptions')\n",
    "plt.xticks(rotation=90)\n",
    "plt.legend()\n",
    "plt.tight_layout()\n",
    "\n",
    "# Show the plot\n",
    "plt.show()\n"
   ],
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ],
      "image/png": "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"
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "execution_count": 5
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-29T21:26:52.369729Z",
     "start_time": "2024-11-29T21:26:51.686887Z"
    }
   },
   "cell_type": "code",
   "source": [
    "import pandas as pd\n",
    "from src.dataloader import DataManager\n",
    "import pathlib\n",
    "\n",
    "path = pathlib.Path(\"data\")\n",
    "df = DataManager.get_misconceptions(path / \"misconception_mapping.csv\")"
   ],
   "id": "c3aab551c9f9d3e4",
   "outputs": [],
   "execution_count": 5
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-29T21:26:54.950150Z",
     "start_time": "2024-11-29T21:26:54.936709Z"
    }
   },
   "cell_type": "code",
   "source": "sum(df.apply(lambda x: len(x['MisconceptionText'].split()*4), axis=1))",
   "id": "9ff767935a8db84b",
   "outputs": [
    {
     "data": {
      "text/plain": [
       "125284"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "execution_count": 6
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-29T21:27:09.183523Z",
     "start_time": "2024-11-29T21:27:09.028149Z"
    }
   },
   "cell_type": "code",
   "source": [
    "import pickle\n",
    "from typing import List\n",
    "import os\n",
    "from openai import OpenAI\n",
    "from dotenv import load_dotenv\n",
    "client = OpenAI(api_key=os.getenv(\"OPENAI_API_KEY\"))\n",
    "\n",
    "load_dotenv()\n",
    "\n",
    "\n",
    "def get_embedding(text, model=\"text-embedding-3-small\"):\n",
    "   text = text.replace(\"\\n\", \" \")\n",
    "   return client.embeddings.create(input = [text], model=model).data[0].embedding"
   ],
   "id": "5213e624a4888f44",
   "outputs": [],
   "execution_count": 7
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-29T21:36:25.182003Z",
     "start_time": "2024-11-29T21:27:27.683240Z"
    }
   },
   "cell_type": "code",
   "source": "df['ada_embedding'] = df.MisconceptionText.apply(lambda x: get_embedding(x))",
   "id": "615eed95ddac65de",
   "outputs": [],
   "execution_count": 8
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-29T21:36:48.132470Z",
     "start_time": "2024-11-29T21:36:45.902634Z"
    }
   },
   "cell_type": "code",
   "source": "df.to_csv(pathlib.Path('data') / \"misconception_mapping_with_embeddings.csv\", index=True)",
   "id": "5628d91edb80e9b9",
   "outputs": [],
   "execution_count": 9
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-29T21:36:55.091751Z",
     "start_time": "2024-11-29T21:36:55.026555Z"
    }
   },
   "cell_type": "code",
   "source": "df.to_pickle(pathlib.Path('data') / \"misconception_mapping_with_embeddings.pkl\")",
   "id": "16767221a5b288c8",
   "outputs": [],
   "execution_count": 10
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-11-29T02:07:38.103931500Z",
     "start_time": "2024-11-29T01:56:12.681211Z"
    }
   },
   "cell_type": "code",
   "source": [
    "# import numpy as np\n",
    "#\n",
    "# df_load = pd.read_csv(pathlib.Path('data') / \"misconception_mapping_with_embeddings.csv\")\n",
    "# df_load['ada_embedding'] = df_load.ada_embedding.apply(eval).apply(np.array)\n"
   ],
   "id": "fa50345c90eb09d7",
   "outputs": [],
   "execution_count": 35
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}