genAi.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Cable', 'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2}, 'execution': {'setup': 'Sit on bench facing away from middle pulley. Place neck in harness cable attachment. Place arms on lower thighs for support.', 'instructions': 'Move head away from pulley by bending neck forward until chin touches upper chest. Return head by hyperextending neck and repeat.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'], 'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']}, 'workout_parameters': {'sets': 4, 'reps': 15, 'bmi_for_exercise': 20.28312872}, 'goals': {'target_audience': 'muscle gain', 'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'female', 'type_of_plan': 'muscle gain'}}\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Lever (plate loaded)', 'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2}, 'execution': {'setup': 'Sit on seat in machine. Position padded lever on face. Position body so neck is in line with lever axis. Grasp handles for support.', 'instructions': 'Move head forward by flexing neck until chin touches upper chest. Return head by hyperextending neck and repeat.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'], 'stabilizer_muscles': ['Latissimus Dorsi', 'Deltoid', 'Posterior', 'Rhomboids', 'Trapezius', 'Middle', 'Trapezius', 'Lower', 'Triceps Brachii', 'Rectus Abdominis', 'Obliques', '']}, 'workout_parameters': {'sets': 3, 'reps': 14, 'bmi_for_exercise': 21.00525585}, 'goals': {'target_audience': 'muscle gain', 'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'female', 'type_of_plan': 'muscle gain'}}\n"
     ]
    }
   ],
   "source": [
    "# import csv\n",
    "\n",
    "# # Load the CSV file and format the data\n",
    "# def process_exercise_csv(csv_path: str) -> list[dict]:\n",
    "#     \"\"\"\n",
    "#     Processes the CSV file to create a structured format similar to the exercise format.\n",
    "\n",
    "#     Parameters:\n",
    "#         csv_path (str): The file path to the CSV file.\n",
    "\n",
    "#     Returns:\n",
    "#         list[dict]: A list of dictionaries with structured data for each exercise.\n",
    "#     \"\"\"\n",
    "#     exercises = []\n",
    "\n",
    "#     with open(csv_path, 'r') as file:\n",
    "#         reader = csv.DictReader(file)\n",
    "\n",
    "#         # Process each row and format the content\n",
    "#         for row in reader:\n",
    "#             exercise_data = {\n",
    "#                 \"exercise\": {\n",
    "#                     \"name\": row.get(\"Exercise Name\", \"N/A\"),\n",
    "#                     \"equipment\": row.get(\"Equipment\", \"N/A\"),\n",
    "#                     \"type\": row.get(\"Utility\", \"N/A\"),\n",
    "#                     \"movement_type\": row.get(\"Mechanics\", \"N/A\"),\n",
    "#                     \"force_type\": row.get(\"Force\", \"N/A\"),\n",
    "#                     \"difficulty\": int(row.get(\"Difficulty (1-5)\", 0))\n",
    "#                 },\n",
    "#                 \"execution\": {\n",
    "#                     \"setup\": row.get(\"Preparation\", \"N/A\"),\n",
    "#                     \"instructions\": row.get(\"Execution\", \"N/A\")\n",
    "#                 },\n",
    "#                 \"muscles\": {\n",
    "#                     \"main_muscle\": row.get(\"Main_muscle\", \"N/A\"),\n",
    "#                     \"target_muscles\": row.get(\"Target_Muscles\", \"\").split(\", \"),\n",
    "#                     \"stabilizer_muscles\": row.get(\"Stabilizer_Muscles\", \"\").split(\", \")\n",
    "#                 },\n",
    "#                 \"workout_parameters\": {\n",
    "#                     \"sets\": int(row.get(\"No_of_Sets\", 0)),\n",
    "#                     \"reps\": int(row.get(\"No_of_Reps\", 0)),\n",
    "#                     \"bmi_for_exercise\": float(row.get(\"BMI\", 0))\n",
    "#                 },\n",
    "#                 \"goals\": {\n",
    "#                     \"target_audience\": \"Muscle Gain\",  # Assuming this is a constant, can be customized\n",
    "#                     \"difficulty_level\": \"Intermediate\" if int(row.get(\"Difficulty (1-5)\", 0)) > 2 else \"Beginner\",\n",
    "#                     \"body_type\": \"Ectomorph\"  # Example constant, can be adjusted based on context\n",
    "#                 }\n",
    "#             }\n",
    "\n",
    "#             # Append each exercise's structured data\n",
    "#             exercises.append(exercise_data)\n",
    "\n",
    "#     return exercises\n",
    "\n",
    "# # Example usage:\n",
    "# csv_path = \"updated_gym_exercise_dataset_men_musclegain.csv\"  # Replace with your actual CSV file path\n",
    "# structured_exercises = process_exercise_csv(csv_path)\n",
    "\n",
    "# # Display the first two exercises' structured content as an example\n",
    "# for exercise in structured_exercises[:2]:\n",
    "#     print(exercise)\n",
    "import csv\n",
    "\n",
    "# Load the CSV file and format the data\n",
    "def process_exercise_csv(csv_path: str) -> list[dict]:\n",
    "    \"\"\"\n",
    "    Processes the CSV file to create a structured format similar to the exercise format,\n",
    "    including gender and type_of_plan (Muscle Gain, Weight Gain, Weight Loss).\n",
    "\n",
    "    Parameters:\n",
    "        csv_path (str): The file path to the CSV file.\n",
    "\n",
    "    Returns:\n",
    "        list[dict]: A list of dictionaries with structured data for each exercise.\n",
    "    \"\"\"\n",
    "    exercises = []\n",
    "\n",
    "    with open(csv_path, 'r') as file:\n",
    "        reader = csv.DictReader(file)\n",
    "\n",
    "        # Process each row and format the content\n",
    "        for row in reader:\n",
    "            exercise_data = {\n",
    "                \"exercise\": {\n",
    "                    \"name\": row.get(\"Exercise Name\", \"N/A\"),\n",
    "                    \"equipment\": row.get(\"Equipment\", \"N/A\"),\n",
    "                    \"type\": row.get(\"Utility\", \"N/A\"),\n",
    "                    \"movement_type\": row.get(\"Mechanics\", \"N/A\"),\n",
    "                    \"force_type\": row.get(\"Force\", \"N/A\"),\n",
    "                    \"difficulty\": int(row.get(\"Difficulty (1-5)\", 0))\n",
    "                },\n",
    "                \"execution\": {\n",
    "                    \"setup\": row.get(\"Preparation\", \"N/A\"),\n",
    "                    \"instructions\": row.get(\"Execution\", \"N/A\")\n",
    "                },\n",
    "                \"muscles\": {\n",
    "                    \"main_muscle\": row.get(\"Main_muscle\", \"N/A\"),\n",
    "                    \"target_muscles\": row.get(\"Target_Muscles\", \"\").split(\", \"),\n",
    "                    \"stabilizer_muscles\": row.get(\"Stabilizer_Muscles\", \"\").split(\", \")\n",
    "                },\n",
    "                \"workout_parameters\": {\n",
    "                    \"sets\": int(row.get(\"No_of_Sets\", 0)),\n",
    "                    \"reps\": int(row.get(\"No_of_Reps\", 0)),\n",
    "                    \"bmi_for_exercise\": float(row.get(\"BMI\", 0))\n",
    "                },\n",
    "                \"goals\": {\n",
    "                    \"target_audience\": row.get(\"type of plan\", \"N/A\"),  # Muscle Gain, Weight Gain, Weight Loss\n",
    "                    \"difficulty_level\": \"Intermediate\" if int(row.get(\"Difficulty (1-5)\", 0)) > 2 else \"Beginner\",\n",
    "                    # Example constant, can be adjusted based on context\n",
    "                },\n",
    "                \"additional_info\": {\n",
    "                    \"gender\": row.get(\"gender\", \"N/A\"),\n",
    "                    \"type_of_plan\": row.get(\"type of plan\", \"N/A\")\n",
    "                }\n",
    "            }\n",
    "\n",
    "            # Append each exercise's structured data\n",
    "            exercises.append(exercise_data)\n",
    "\n",
    "    return exercises\n",
    "\n",
    "# Example usage:\n",
    "csv_path = \"GYM_File.csv\"  # Replace with your actual CSV file path\n",
    "structured_exercises = process_exercise_csv(csv_path)\n",
    "\n",
    "# Display the first two exercises' structured content as an example\n",
    "for exercise in structured_exercises[:2]:\n",
    "    print(exercise)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4a8ad7607ad3409f8382a4b7e4c57ac5",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "0it [00:00, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from tqdm.auto import tqdm # for progress bars, requires !pip install tqdm \n",
    "import json\n",
    "\n",
    "\n",
    "\n",
    "# Open PDF and get lines/pages\n",
    "# Note: this only focuses on text, rather than images/figures etc\n",
    "def open_and_read_pdf(list: str) -> list[dict]:\n",
    "    \"\"\"\n",
    "    Opens a PDF file, reads its text content page by page, and collects statistics.\n",
    "\n",
    "    Parameters:\n",
    "        pdf_path (str): The file path to the PDF document to be opened and read.\n",
    "\n",
    "    Returns:\n",
    "        list[dict]: A list of dictionaries, each containing the page number\n",
    "        (adjusted), character count, word count, sentence count, token count, and the extracted text\n",
    "        for each page.\n",
    "    \"\"\"\n",
    "    json_text = json.dumps(list)\n",
    "    page_char_count = len(json_text)\n",
    "\n",
    "    # Calculate page token count\n",
    "    page_token_count = page_char_count / 4 \n",
    "    pages_and_texts = []\n",
    "    for page_number, page in tqdm(enumerate(list)):  # iterate the document pages\n",
    "        text = page # get plain text encoded as UTF-8\n",
    "        pages_and_texts.append({\"page_number\": page_number ,  # adjust page numbers since our PDF starts on page 42\n",
    "                                \"page_char_count\": len(text),\n",
    "                                \"page_token_count\": len(text) / 4,  # 1 token = ~4 chars, see: https://help.openai.com/en/articles/4936856-what-are-tokens-and-how-to-count-them\n",
    "                                \"text\": text})\n",
    "    return pages_and_texts\n",
    "\n",
    "pages_and_texts = open_and_read_pdf(structured_exercises)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>page_number</th>\n",
       "      <th>page_char_count</th>\n",
       "      <th>page_token_count</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Neck Flexion', 'equipme...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Neck Flexion', 'equipme...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Lateral Neck Flexion', ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Neck Flexion', 'equipme...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Lateral Neck Flexion', ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   page_number  page_char_count  page_token_count  \\\n",
       "0            0                6               1.5   \n",
       "1            1                6               1.5   \n",
       "2            2                6               1.5   \n",
       "3            3                6               1.5   \n",
       "4            4                6               1.5   \n",
       "\n",
       "                                                text  \n",
       "0  {'exercise': {'name': 'Neck Flexion', 'equipme...  \n",
       "1  {'exercise': {'name': 'Neck Flexion', 'equipme...  \n",
       "2  {'exercise': {'name': 'Lateral Neck Flexion', ...  \n",
       "3  {'exercise': {'name': 'Neck Flexion', 'equipme...  \n",
       "4  {'exercise': {'name': 'Lateral Neck Flexion', ...  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.DataFrame(pages_and_texts)\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: sentence-transformers in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (3.0.1)\n",
      "Requirement already satisfied: transformers<5.0.0,>=4.34.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sentence-transformers) (4.42.3)\n",
      "Requirement already satisfied: tqdm in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sentence-transformers) (4.66.4)\n",
      "Requirement already satisfied: torch>=1.11.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sentence-transformers) (2.2.1+cu121)\n",
      "Requirement already satisfied: numpy in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sentence-transformers) (1.26.4)\n",
      "Requirement already satisfied: scikit-learn in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sentence-transformers) (1.3.2)\n",
      "Requirement already satisfied: scipy in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sentence-transformers) (1.11.4)\n",
      "Requirement already satisfied: huggingface-hub>=0.15.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sentence-transformers) (0.23.4)\n",
      "Requirement already satisfied: Pillow in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sentence-transformers) (10.4.0)\n",
      "Requirement already satisfied: filelock in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub>=0.15.1->sentence-transformers) (3.15.4)\n",
      "Requirement already satisfied: fsspec>=2023.5.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub>=0.15.1->sentence-transformers) (2024.6.1)\n",
      "Requirement already satisfied: packaging>=20.9 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub>=0.15.1->sentence-transformers) (24.1)\n",
      "Requirement already satisfied: pyyaml>=5.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub>=0.15.1->sentence-transformers) (6.0.1)\n",
      "Requirement already satisfied: requests in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub>=0.15.1->sentence-transformers) (2.32.3)\n",
      "Requirement already satisfied: typing-extensions>=3.7.4.3 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub>=0.15.1->sentence-transformers) (4.12.2)\n",
      "Requirement already satisfied: sympy in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (1.12.1)\n",
      "Requirement already satisfied: networkx in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (3.3)\n",
      "Requirement already satisfied: jinja2 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (3.1.4)\n",
      "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cudnn-cu12==8.9.2.26 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (8.9.2.26)\n",
      "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (12.1.3.1)\n",
      "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (11.0.2.54)\n",
      "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (10.3.2.106)\n",
      "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (11.4.5.107)\n",
      "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (12.1.0.106)\n",
      "Requirement already satisfied: nvidia-nccl-cu12==2.19.3 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (2.19.3)\n",
      "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (12.1.105)\n",
      "Requirement already satisfied: triton==2.2.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch>=1.11.0->sentence-transformers) (2.2.0)\n",
      "Requirement already satisfied: nvidia-nvjitlink-cu12 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch>=1.11.0->sentence-transformers) (12.5.82)\n",
      "Requirement already satisfied: regex!=2019.12.17 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from transformers<5.0.0,>=4.34.0->sentence-transformers) (2024.5.15)\n",
      "Requirement already satisfied: safetensors>=0.4.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from transformers<5.0.0,>=4.34.0->sentence-transformers) (0.4.3)\n",
      "Requirement already satisfied: tokenizers<0.20,>=0.19 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from transformers<5.0.0,>=4.34.0->sentence-transformers) (0.19.1)\n",
      "Requirement already satisfied: joblib>=1.1.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from scikit-learn->sentence-transformers) (1.4.2)\n",
      "Requirement already satisfied: threadpoolctl>=2.0.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from scikit-learn->sentence-transformers) (3.5.0)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from jinja2->torch>=1.11.0->sentence-transformers) (2.1.5)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests->huggingface-hub>=0.15.1->sentence-transformers) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests->huggingface-hub>=0.15.1->sentence-transformers) (3.7)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests->huggingface-hub>=0.15.1->sentence-transformers) (2.2.2)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests->huggingface-hub>=0.15.1->sentence-transformers) (2024.7.4)\n",
      "Requirement already satisfied: mpmath<1.4.0,>=1.1.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sympy->torch>=1.11.0->sentence-transformers) (1.3.0)\n",
      "\n",
      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.1.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.2\u001b[0m\n",
      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "%pip install sentence-transformers\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "    Hello world!\n",
      "[ 1.91737153e-02  2.87365373e-02 -1.23540815e-02  1.58220939e-02\n",
      "  7.90899619e-02 -9.76034440e-03  7.49561517e-03  5.52258082e-02\n",
      "  1.88754424e-02 -2.63798703e-02 -2.68068500e-02 -3.33473235e-02\n",
      " -3.00286710e-02  3.89383473e-02  7.69484565e-02 -7.68074542e-02\n",
      "  6.97796494e-02 -6.93720859e-03 -4.48980965e-02  1.21879680e-02\n",
      "  1.03991563e-02  8.78110062e-03  1.08342348e-02  5.92033267e-02\n",
      "  1.70315858e-02 -2.56328769e-02  9.10411682e-03  3.73560283e-03\n",
      "  3.32366712e-02  5.40160621e-03 -3.08646597e-02  5.92166372e-03\n",
      "  5.06224595e-02  6.56094775e-02  2.12388136e-06 -5.11630215e-02\n",
      "  2.44774185e-02 -7.60708330e-03 -7.54034705e-03  3.69985867e-03\n",
      " -2.33552675e-03  2.81633567e-02 -1.65304970e-02  9.42942686e-03\n",
      "  6.09955145e-03 -4.32090759e-02 -5.53827675e-04 -7.96658080e-03\n",
      "  1.98367257e-02  2.02401243e-02 -4.93253069e-03  1.22366399e-02\n",
      " -6.17669933e-02 -2.02074158e-03  4.50151749e-02  3.01424600e-02\n",
      "  3.64244357e-02  1.64621267e-02  1.17515689e-02 -4.62706275e-02\n",
      "  5.13225719e-02 -3.50427032e-02 -2.48365179e-02 -6.78302627e-03\n",
      " -6.15483616e-04  3.80288474e-02 -2.47982144e-02  3.53081450e-02\n",
      " -7.37571903e-03  6.03849217e-02  3.16642448e-02 -2.13077981e-02\n",
      "  3.91404666e-02  7.03454688e-02  1.12680495e-02 -6.77807108e-02\n",
      " -3.48133594e-02  7.59453103e-02  2.43842732e-02  1.01228049e-02\n",
      " -2.56590825e-02  1.57113466e-02 -1.00360923e-02 -2.50009038e-02\n",
      " -2.85806153e-02  7.33537152e-02  1.09247593e-02 -6.60133827e-03\n",
      "  2.30301060e-02  2.76657622e-02 -1.60924215e-02 -9.57247894e-03\n",
      " -2.69190744e-02  2.79842615e-02  5.70767326e-03 -4.97548729e-02\n",
      "  2.81997751e-02  1.87633634e-02  6.00257590e-02 -6.98942989e-02\n",
      "  5.58036333e-03  4.47613671e-02  1.92689765e-02  1.83421634e-02\n",
      "  1.19073680e-02  2.95258127e-02  1.27076041e-02 -9.85340625e-02\n",
      " -4.09881491e-03 -6.13117078e-03 -3.39137577e-02  1.65321361e-02\n",
      "  5.16281910e-02  1.97649803e-02 -4.58239503e-02 -5.16074651e-04\n",
      "  1.53688416e-02  4.48495336e-02  1.93863735e-02 -4.05021664e-03\n",
      " -6.72021834e-03 -3.42007517e-03  1.66347772e-02  8.94476473e-03\n",
      "  1.62143484e-02 -1.00794900e-02 -2.36646868e-02  1.02804287e-03\n",
      " -4.50739339e-02 -7.11200237e-02 -7.62604736e-03  1.50395313e-03\n",
      "  3.16536240e-02  1.31290266e-02  1.94331557e-02  3.16213034e-02\n",
      " -3.70885283e-02  1.17648812e-02  3.78226042e-02 -3.26601416e-03\n",
      " -3.49018052e-02  1.35727106e-02  1.16819767e-02 -1.83340702e-02\n",
      " -3.90327274e-04 -3.67906236e-04  5.92751279e-02  7.33461604e-02\n",
      "  1.57829020e-02 -1.74836181e-02 -1.02733038e-02 -1.00880936e-02\n",
      " -9.01498049e-02  2.13089189e-03  4.13726084e-02  4.44958061e-02\n",
      "  1.84938237e-02 -3.75402570e-02 -8.33088756e-02 -4.01758179e-02\n",
      " -9.14124399e-03  1.78720672e-02  3.49662043e-02  4.31776559e-03\n",
      "  4.44156341e-02  3.12555619e-02  1.65182762e-02 -3.32650431e-02\n",
      "  1.21722382e-03 -1.08136781e-01 -1.71430837e-02 -7.12407660e-03\n",
      " -3.86400032e-03  2.97890808e-02 -4.40895883e-03 -4.46760692e-02\n",
      " -8.18911940e-02  4.49774191e-02  3.11202072e-02  4.32216882e-04\n",
      " -3.04392371e-02 -9.11770463e-02 -2.08065379e-02 -3.35289352e-02\n",
      " -4.90985326e-02  4.62109037e-02  2.62859706e-02  1.62741952e-02\n",
      " -3.98702137e-02  3.70658822e-02  1.64647233e-02  2.29480751e-02\n",
      " -4.77646776e-02 -2.73194835e-02 -3.84584740e-02  1.27943289e-02\n",
      "  1.61465909e-02 -2.51712110e-02  3.58927697e-02 -8.60399231e-02\n",
      " -2.51330920e-02  5.85450083e-02  1.88994538e-02 -1.10534495e-02\n",
      " -2.88860183e-02  1.97064541e-02  7.14479461e-02 -9.70275234e-03\n",
      "  3.19969356e-02  6.47671968e-02  2.58722692e-03  3.74229113e-03\n",
      "  2.30630115e-02  5.59966452e-02 -5.46833267e-03  8.28381721e-03\n",
      "  1.48229245e-02  6.77184537e-02 -2.22103484e-02  9.26544890e-02\n",
      "  3.09276744e-04  1.13492999e-02  4.54021767e-02 -2.84848604e-02\n",
      "  3.68759036e-02 -5.37904315e-02  1.15620615e-02  2.63585616e-02\n",
      " -3.00341696e-02  2.20608059e-02 -3.82237472e-02 -1.32851526e-02\n",
      "  1.17402999e-02 -3.87181230e-02 -2.11313404e-02 -2.12081559e-02\n",
      "  6.13836497e-02 -2.09866166e-02  4.60760109e-02 -1.40556768e-01\n",
      "  3.14407460e-02  3.47950459e-02  1.34149464e-02 -3.90576869e-02\n",
      "  5.26755750e-02  2.82531884e-02 -7.78555684e-03  2.50696428e-02\n",
      " -1.64742395e-02  1.05087189e-02  1.58723406e-02  3.98026314e-03\n",
      " -9.53608453e-02 -1.39455935e-02 -2.06975522e-03 -4.27396707e-02\n",
      " -4.07640375e-02  2.21871189e-03 -3.38128284e-02 -4.63380702e-02\n",
      "  2.03128066e-02 -3.97231914e-02  2.36008619e-03  3.08819879e-02\n",
      "  4.42393869e-02  2.09606364e-02  3.72628169e-03 -6.12542108e-02\n",
      " -2.98867635e-02 -4.14774287e-03  2.18838174e-02  2.74256356e-02\n",
      "  1.80789232e-02 -4.87662246e-03 -3.81266326e-02 -1.12980939e-02\n",
      "  4.82324474e-02  5.27306367e-03  5.80645213e-03  3.92983742e-02\n",
      "  3.25348675e-02 -8.19188915e-03 -3.23430263e-02  1.33359013e-02\n",
      "  1.63216481e-03  3.75178270e-02  4.17793952e-02 -3.35737281e-02\n",
      "  3.94029580e-02  3.38618923e-03 -4.91535179e-02  5.30127843e-04\n",
      " -3.68258357e-02 -2.79936027e-02  1.96319781e-02 -2.28686035e-02\n",
      "  1.09326849e-02 -6.48552775e-02 -1.26678161e-02 -2.52633654e-02\n",
      " -3.43988761e-02  3.62152606e-02  2.49012653e-02 -1.61011107e-02\n",
      "  3.05041373e-02  7.52810575e-03  1.22987134e-02 -6.00183718e-02\n",
      " -2.96193343e-02  5.00400811e-02  3.96712907e-02 -6.70813723e-03\n",
      " -1.35938413e-02  3.60015891e-02  1.72831602e-02 -1.03489999e-02\n",
      " -3.32857780e-02 -9.80150327e-02  2.04230007e-02 -1.96092948e-02\n",
      "  1.21857068e-02  3.33469063e-02 -2.07348112e-02  2.30167657e-02\n",
      " -4.36715409e-02  8.60219076e-02  2.86375708e-03  4.26322632e-02\n",
      "  4.50773314e-02  9.71542951e-03  2.12640744e-02  1.51656084e-02\n",
      " -2.43581701e-02 -5.37792780e-03 -4.28857282e-02 -5.80270141e-02\n",
      "  2.51577161e-02 -4.96683344e-02 -1.73642971e-02  1.64442956e-02\n",
      " -4.84491736e-02 -5.09288050e-02 -3.81456874e-02  8.45364574e-03\n",
      " -4.36721332e-02 -3.15041319e-02  7.95564801e-03 -2.85016187e-02\n",
      " -7.62111843e-02 -1.97426248e-02 -3.38924229e-02 -5.05803451e-02\n",
      "  1.56861953e-02 -1.03990762e-02 -1.30822053e-02 -3.45803574e-02\n",
      " -1.23006031e-02  8.99054948e-03 -2.08482873e-02 -3.54249291e-02\n",
      " -1.34092551e-02 -4.39251810e-02  3.70026380e-02  8.23191330e-02\n",
      " -6.59834826e-04 -4.02661189e-02  3.09138130e-02  4.47829738e-02\n",
      " -4.74218801e-02 -8.96502659e-03 -6.94046635e-03  4.97259386e-02\n",
      " -1.87434796e-02  5.94354421e-02 -2.65647285e-02  2.17275731e-02\n",
      " -4.95409593e-02 -1.01867728e-02  2.88019758e-02 -9.03270859e-03\n",
      "  3.82599123e-02 -4.37173154e-03  4.64569256e-02 -7.70743564e-03\n",
      " -2.65655201e-02 -2.91740354e-02 -1.09978439e-02  1.44797415e-02\n",
      "  4.97333659e-03  5.80650903e-02  2.29797475e-02 -7.41408393e-02\n",
      "  1.89360455e-02  8.61386955e-03  1.79641612e-03  3.41153890e-02\n",
      " -1.21506186e-04  9.74259712e-03 -4.22965214e-02  2.68055927e-02\n",
      " -2.85971351e-02  2.09478643e-02  7.28382990e-02  2.41208170e-02\n",
      "  6.19544536e-02 -5.87005168e-03  7.47879893e-02  1.24215158e-02\n",
      "  1.19447522e-02 -1.68880913e-03 -3.58085185e-02  1.30939181e-03\n",
      "  1.37497333e-03 -3.13581713e-02 -1.62730869e-02 -6.01773225e-02\n",
      " -2.37561800e-02 -6.51233271e-02 -2.40334626e-02 -9.39350296e-03\n",
      "  1.07824737e-02 -4.09287773e-03  3.66987922e-04  4.73605394e-02\n",
      " -3.64044448e-04  6.08529057e-03  2.88276002e-02  1.13299571e-01\n",
      "  1.95047259e-02  1.21034896e-02 -9.62612499e-03  1.03835491e-02\n",
      "  5.67056425e-02  5.13534695e-02 -6.35669800e-03 -1.92079097e-02\n",
      "  6.95613399e-03 -6.90898020e-03 -6.09237477e-02 -2.48148255e-02\n",
      " -6.31580641e-03 -7.14286743e-03 -1.27549088e-02 -8.85819551e-04\n",
      " -1.93067733e-02  4.31431346e-02  1.02687357e-02 -3.81555632e-02\n",
      "  2.70078946e-02 -7.15295970e-02  2.45865360e-02 -1.61340088e-02\n",
      " -1.77070796e-02 -6.26399443e-02  1.64759122e-02 -7.19615445e-02\n",
      "  8.69675772e-04  2.96783121e-03  3.94573510e-02 -4.71781008e-02\n",
      "  8.78058672e-02  2.12599132e-02 -6.53822646e-02  1.24869458e-02\n",
      "  5.18307053e-02 -1.11388240e-03 -6.35768753e-03  2.98758931e-02\n",
      "  5.15517332e-02 -2.73086270e-03 -5.80668040e-02  6.33830344e-03\n",
      " -2.52553355e-02 -2.79888380e-02 -1.92381181e-02  5.31651862e-02\n",
      " -6.86165085e-03 -4.61082831e-02 -7.04449192e-02 -2.20946912e-02\n",
      " -7.25357619e-04 -1.82642844e-02  8.22450686e-03  5.05150901e-03\n",
      " -2.08496712e-02 -5.82599528e-02  5.17655984e-02 -1.64420493e-02\n",
      " -3.70764136e-02  1.87967252e-02 -1.42324837e-02  8.72096047e-04\n",
      "  6.54906332e-02  3.16015929e-02 -6.53313175e-02  3.46708558e-02\n",
      "  3.84957194e-02 -4.14432585e-02 -1.08024413e-02  4.43097763e-02\n",
      "  5.03093237e-03  3.02684400e-02 -2.50593163e-02 -1.86341733e-03\n",
      "  2.69365087e-02 -9.18115675e-03  1.42915407e-03  2.68606166e-03\n",
      "  1.12632476e-02 -4.95557152e-02  1.31691955e-02  2.07001697e-02\n",
      " -4.18449752e-02 -8.94659534e-02  2.33849715e-02  9.91613492e-02\n",
      " -6.09314826e-04  1.82036981e-02  2.11171284e-02 -5.05482815e-02\n",
      "  3.40849422e-02 -1.32596679e-02 -2.13379636e-02 -1.95383616e-02\n",
      "  2.38800328e-02  1.12978276e-02 -4.00543446e-03 -2.41861306e-02\n",
      "  1.32577326e-02  2.26723147e-03 -5.02162576e-02 -1.11844484e-03\n",
      "  4.78284173e-02  1.90947298e-02 -2.90436614e-02  5.86056747e-02\n",
      "  5.28842919e-02  1.92839820e-02 -3.04641053e-02  8.17226693e-02\n",
      " -2.49548294e-02  3.35669331e-02 -1.17796725e-02  2.69153286e-02\n",
      " -2.51451903e-03  6.32932410e-02  3.30374315e-02 -3.00421845e-03\n",
      " -1.66062005e-02 -3.92380245e-02  3.08183068e-03  2.80030780e-02\n",
      " -8.10901914e-03 -9.33826063e-03 -1.11555951e-02 -7.47020709e-33\n",
      " -3.22055444e-02  3.56099457e-02 -5.82496412e-02 -1.60795357e-02\n",
      " -6.52420297e-02 -3.33212726e-02  2.13987883e-02  1.50862718e-02\n",
      " -1.73138846e-02  4.09015864e-02 -3.87609899e-02  1.05188992e-02\n",
      "  2.50659175e-02  3.13543645e-03 -2.31369585e-02 -4.52400884e-03\n",
      "  3.57131138e-02  4.11871746e-02 -2.86745548e-04  1.74105614e-02\n",
      "  1.22149214e-02  2.46188808e-02  4.59370650e-02 -1.33532612e-02\n",
      "  7.13071525e-02 -2.96959141e-03 -3.59614864e-02 -2.72705145e-02\n",
      "  6.44440949e-02 -1.71523988e-02 -5.55020683e-02  8.80477764e-03\n",
      " -1.35961948e-02  1.90238785e-02 -5.36696939e-03  3.12593989e-02\n",
      " -1.66032538e-02 -1.22872731e-02  5.07901683e-02 -1.03928121e-02\n",
      "  2.11503115e-02  4.54903580e-02 -4.67937365e-02 -4.98828366e-02\n",
      "  1.74909178e-02  3.21024917e-02  4.09138622e-03  2.73100119e-02\n",
      "  3.92665677e-02 -1.01147238e-02 -7.32675344e-02 -7.26894289e-02\n",
      " -1.01327011e-02 -6.41313866e-02 -1.01923160e-01  7.00137671e-03\n",
      " -5.76170534e-03 -6.46813819e-03  2.20476966e-02  1.01690292e-02\n",
      "  6.98977411e-02 -4.26165164e-02 -4.28642109e-02  4.30842862e-02\n",
      "  1.58814657e-02  7.93900937e-02  1.16006128e-01  2.20975280e-02\n",
      " -3.04972921e-02 -7.12647513e-02 -3.90832089e-02 -3.21927085e-03\n",
      " -4.58594374e-02  4.95483093e-02 -3.22161429e-02 -9.35604970e-04\n",
      " -6.84287921e-02 -3.27659734e-02 -7.94430170e-03 -3.21570784e-02\n",
      " -3.96366119e-02 -1.84605643e-02 -1.14914216e-02  2.94621494e-02\n",
      "  1.94461886e-02  4.15210575e-02 -1.90555688e-03  2.43718121e-02\n",
      " -5.03088720e-02  1.94552094e-02 -2.26781946e-02 -6.72485754e-02\n",
      "  2.77109202e-02 -5.28804250e-02 -1.71071012e-02 -2.00080313e-02\n",
      "  4.75358218e-02 -1.16792498e-02 -2.23242352e-03 -1.98743343e-02\n",
      "  3.11619546e-02 -6.44603046e-03 -5.14813773e-02  3.10154650e-02\n",
      "  4.21763398e-03  5.90362251e-02 -5.95617741e-02 -1.20727355e-02\n",
      " -5.79645820e-02 -1.52969221e-02  1.19450176e-02  9.56446398e-04\n",
      " -7.11018103e-04  9.00592934e-03 -1.07602952e-02  4.16445825e-03\n",
      "  1.98662113e-02  1.48234877e-03 -4.13043611e-02 -2.06007268e-02\n",
      "  3.54354493e-02 -2.52422038e-02 -4.03880961e-02  1.35236662e-02\n",
      " -3.02803759e-02 -1.57543514e-02 -6.40296610e-03  2.88539771e-02\n",
      "  3.77058983e-02 -1.57546159e-02  1.93659272e-02 -3.18935812e-02\n",
      "  2.74357859e-07  1.36573752e-02 -2.96957642e-02 -1.86281409e-02\n",
      "  4.90262844e-02  1.95596181e-02  6.26194756e-03 -2.18237136e-02\n",
      "  7.88233727e-02  5.18120117e-02  1.02162333e-02  1.23885563e-02\n",
      " -7.14070629e-03 -1.12918392e-02 -7.13305473e-02 -5.56696467e-02\n",
      "  5.20570390e-03 -5.79464845e-02 -1.12188607e-01 -2.72846259e-02\n",
      "  1.57685000e-02 -1.67696122e-02  5.12122596e-03 -4.01668437e-02\n",
      " -1.31391850e-03  2.60912944e-02 -2.37525515e-02  3.94153148e-02\n",
      " -3.82065028e-03 -6.60820771e-03  9.01799742e-03 -3.99146192e-02\n",
      " -4.71640788e-02  2.90916674e-02 -3.72883566e-02  1.84368473e-02\n",
      " -2.63714921e-02  1.30457794e-02  2.67634634e-02  6.50384231e-03\n",
      "  8.21977481e-02 -3.83656211e-02  3.92347723e-02 -2.39917226e-02\n",
      " -3.60896662e-02  1.78224556e-02  1.22119403e-02  3.41273658e-02\n",
      "  8.46597552e-03 -2.43397970e-02 -4.21196520e-02 -1.19033698e-02\n",
      " -2.82946341e-02  7.91271962e-03  1.75073128e-02 -1.20986598e-02\n",
      "  2.61467937e-02  6.27233833e-02  2.33795103e-02  3.18712294e-02\n",
      " -4.35521975e-02  2.96295155e-03 -1.10605076e-01 -5.36747044e-03\n",
      "  2.03250051e-02 -8.06444976e-03 -1.79817062e-02 -2.39730012e-02\n",
      "  1.15365093e-34 -1.68201923e-02  2.72668395e-02  2.62050256e-02\n",
      "  6.68588802e-02 -1.96504463e-02 -9.88802291e-04 -4.00086492e-02\n",
      "  1.52709624e-02  4.57853265e-02 -3.14779654e-02 -1.91930011e-02]\n"
     ]
    }
   ],
   "source": [
    "from sentence_transformers import SentenceTransformer\n",
    "device = \"cuda\" if torch.cuda.is_available() else \"cpu\"\n",
    "model=SentenceTransformer(model_name_or_path=\"all-mpnet-base-v2\", \n",
    "                                      device=device)\n",
    "sentences=[\n",
    "\"    Hello world!\"\n",
    "]\n",
    "embeddings=model.encode(sentences)\n",
    "embeddings_dist=dict(zip(sentences,embeddings))\n",
    "for sentence, embedding in embeddings_dist.items():\n",
    "  print(sentence)\n",
    "  print(embedding)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "embeddings_text=pages_and_texts"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "for item in embeddings_text:\n",
    "  text_to_embed = json.dumps(item['text'])\n",
    "  item[\"embedding\"]=model.encode(text_to_embed,convert_to_tensor=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "# import numpy as np\n",
    "# import pandas as pd\n",
    "\n",
    "# # Assuming 'text_chunk_embeddings' is a tensor\n",
    "# embeddings_text_np = embeddings_text.numpy()  # Convert PyTorch tensor to NumPy array\n",
    "\n",
    "# # Save embeddings to CSV\n",
    "# np.savetxt('text_chunk_embeddings.csv', embeddings_text_np, delimiter=',')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'page_number': 0,\n",
       " 'page_char_count': 6,\n",
       " 'page_token_count': 1.5,\n",
       " 'text': {'exercise': {'name': 'Neck Flexion',\n",
       "   'equipment': 'Cable',\n",
       "   'type': 'Basic or Auxiliary',\n",
       "   'movement_type': 'Isolated',\n",
       "   'force_type': 'Pull',\n",
       "   'difficulty': 2},\n",
       "  'execution': {'setup': 'Sit on bench facing away from middle pulley. Place neck in harness cable attachment. Place arms on lower thighs for support.',\n",
       "   'instructions': 'Move head away from pulley by bending neck forward until chin touches upper chest. Return head by hyperextending neck and repeat.'},\n",
       "  'muscles': {'main_muscle': 'Neck',\n",
       "   'target_muscles': ['Sternocleidomastoid,'],\n",
       "   'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
       "  'workout_parameters': {'sets': 4,\n",
       "   'reps': 15,\n",
       "   'bmi_for_exercise': 20.28312872},\n",
       "  'goals': {'target_audience': 'muscle gain', 'difficulty_level': 'Beginner'},\n",
       "  'additional_info': {'gender': 'female', 'type_of_plan': 'muscle gain'}},\n",
       " 'embedding': tensor([ 3.0419e-02, -6.3200e-03, -1.0122e-02,  2.1678e-02, -7.6340e-03,\n",
       "          3.6913e-02, -8.5636e-03,  4.1395e-02, -2.3450e-02, -3.4495e-04,\n",
       "          3.8711e-02,  4.6405e-02,  6.3204e-02,  3.9911e-03, -6.5903e-03,\n",
       "         -6.2610e-02, -1.9130e-02,  2.5541e-02,  5.6282e-02, -1.1788e-02,\n",
       "         -1.8855e-02,  3.5343e-02, -1.2520e-02, -3.5850e-02, -6.0564e-03,\n",
       "         -2.7441e-02, -5.8523e-02,  3.0393e-02, -8.9362e-03, -9.7978e-02,\n",
       "          1.7077e-02,  1.3824e-03,  2.5907e-02, -3.5247e-02,  2.0264e-06,\n",
       "         -3.4333e-02, -2.2077e-03,  5.7767e-03, -4.8866e-02,  1.7982e-02,\n",
       "         -4.1097e-03,  3.9733e-02,  1.4877e-02, -2.1072e-02,  1.5585e-02,\n",
       "         -3.6810e-02, -8.6576e-03,  1.4204e-02,  3.5090e-02,  7.6708e-02,\n",
       "         -1.3828e-02,  1.5437e-03, -1.9630e-02, -3.1267e-02,  6.3099e-02,\n",
       "         -5.7229e-02, -1.4906e-02,  3.5591e-02,  7.1930e-03,  8.0838e-02,\n",
       "         -3.1209e-02, -1.7498e-02, -1.2694e-02,  6.4387e-02,  1.4251e-02,\n",
       "          3.9119e-03, -9.9548e-03, -9.9104e-02,  1.5691e-02,  4.0052e-02,\n",
       "          6.3073e-02,  4.1892e-02, -9.2683e-03,  2.7157e-02,  3.9637e-02,\n",
       "         -2.0065e-02, -3.0162e-02,  1.4630e-02,  3.2373e-02, -7.1435e-03,\n",
       "         -2.4071e-02,  5.7933e-03, -2.3918e-02, -4.1277e-02,  3.3539e-02,\n",
       "         -7.8887e-03, -2.5252e-02, -1.5436e-02,  1.4374e-02, -4.5505e-02,\n",
       "          2.7738e-02, -2.1217e-02,  3.1044e-02,  1.9587e-02, -2.6732e-02,\n",
       "         -2.1790e-02,  3.0179e-04, -1.7002e-01,  5.6018e-02, -2.3925e-03,\n",
       "          2.7984e-02, -1.1880e-02,  8.4809e-02,  7.7953e-03, -4.3009e-03,\n",
       "          2.7990e-02, -1.2276e-04,  6.4424e-03, -5.2192e-02,  7.5311e-02,\n",
       "         -4.1986e-02,  3.3046e-02, -5.8013e-02, -1.7621e-02,  1.1469e-02,\n",
       "          5.3172e-03, -4.3476e-02,  1.1263e-02,  2.9692e-02, -3.1192e-03,\n",
       "          6.8446e-03,  4.6512e-02, -1.5275e-02,  8.1101e-03, -3.1887e-02,\n",
       "         -5.1949e-02, -6.1584e-03,  1.1960e-02,  2.1210e-02,  1.6497e-02,\n",
       "         -2.9342e-02,  4.8622e-02,  2.0913e-02, -4.5065e-02,  1.7807e-03,\n",
       "         -4.6591e-02, -1.1726e-02,  1.3999e-03,  4.8896e-02, -5.6764e-02,\n",
       "          4.8559e-02, -1.5819e-02, -2.3102e-02, -1.5622e-02,  2.2940e-02,\n",
       "         -1.8686e-02,  2.8966e-03,  2.8234e-02, -1.8647e-02,  2.5194e-03,\n",
       "         -9.4499e-02,  9.3581e-03,  4.3300e-03,  5.6499e-02,  2.1107e-02,\n",
       "         -1.1030e-02, -2.0775e-02,  2.0658e-03, -2.8581e-02,  5.8473e-02,\n",
       "          2.6830e-02,  3.4429e-02, -4.2053e-02, -2.9459e-02,  8.7228e-03,\n",
       "         -7.0686e-03, -1.8374e-02, -4.0635e-02,  1.8291e-02,  2.2386e-02,\n",
       "         -3.8888e-02, -6.9852e-03, -2.3043e-02,  9.3983e-02, -5.5753e-03,\n",
       "          1.0572e-01,  8.8439e-02, -7.2146e-02, -6.1434e-02,  1.5526e-02,\n",
       "         -1.8533e-02, -1.0470e-01, -2.0899e-02, -6.3128e-02,  9.8377e-03,\n",
       "          3.4745e-02, -4.9232e-02,  4.4976e-02, -3.8421e-02,  5.8412e-03,\n",
       "         -1.7977e-02, -3.0612e-02, -3.2372e-02, -2.5047e-02,  6.5533e-02,\n",
       "          5.1177e-02, -3.9441e-02,  2.4125e-02, -2.6500e-02, -2.2684e-02,\n",
       "          1.7797e-02,  5.0208e-03, -2.2185e-02, -6.2069e-02, -1.5281e-03,\n",
       "          1.6806e-02,  1.5359e-03,  2.5444e-02, -2.7088e-02,  1.4376e-02,\n",
       "          4.5612e-03,  2.9052e-02, -3.9888e-02, -9.6707e-03, -3.4232e-02,\n",
       "          6.1579e-02, -5.8069e-03, -6.2691e-02, -1.8078e-02, -1.3788e-02,\n",
       "         -7.4246e-04,  2.1923e-02, -4.2760e-02,  2.7040e-03, -9.7597e-02,\n",
       "          6.9708e-03, -3.7123e-03, -1.7816e-02,  4.0723e-02,  1.4907e-02,\n",
       "          2.9866e-02,  1.7118e-03,  2.1276e-02, -2.1829e-02, -4.3593e-02,\n",
       "         -6.4045e-04,  3.7722e-02, -3.7175e-04,  5.0073e-02, -3.4802e-02,\n",
       "         -1.5121e-02,  6.8778e-03,  1.5362e-02, -3.5831e-02,  4.1319e-02,\n",
       "         -1.0121e-02,  5.8459e-03,  2.0843e-02,  1.0021e-02,  2.8672e-03,\n",
       "         -2.0533e-02,  1.6375e-02, -1.8107e-02,  2.0932e-02, -5.7535e-02,\n",
       "         -1.2505e-02, -4.9204e-02, -1.1710e-02, -8.1004e-02, -2.7131e-02,\n",
       "          9.2549e-02, -2.1157e-02, -1.3159e-02,  1.5873e-02,  8.2322e-02,\n",
       "          3.8830e-03,  4.5700e-03, -2.8903e-03,  3.1329e-02, -1.6312e-02,\n",
       "          3.7105e-02,  1.1444e-02,  1.4607e-02, -1.8036e-02, -1.9538e-02,\n",
       "         -3.3014e-02,  2.5229e-02, -5.6220e-02, -2.2970e-02, -3.6099e-02,\n",
       "          1.9638e-02,  7.3662e-03, -1.8633e-02,  5.1218e-03, -1.9978e-02,\n",
       "          2.8361e-02,  2.0527e-02,  2.6693e-02,  7.4735e-03, -1.6280e-02,\n",
       "          4.7863e-02,  2.4420e-02,  1.7694e-02, -8.1150e-02,  3.2683e-02,\n",
       "          2.5339e-02,  3.6672e-02, -1.0882e-02,  2.6050e-02, -1.6848e-02,\n",
       "          1.5959e-02,  7.6366e-02, -5.3470e-03,  1.8950e-02, -9.1476e-02,\n",
       "         -4.7661e-02,  2.8459e-04, -9.0595e-02, -3.7578e-03,  1.3253e-03,\n",
       "          6.2020e-02,  1.3645e-03, -3.1600e-02, -3.6997e-02,  2.4806e-02,\n",
       "          3.4978e-03, -1.0917e-02, -2.0881e-02, -6.8185e-03,  4.7053e-02,\n",
       "         -4.2596e-03,  3.6719e-03, -1.0026e-02,  1.9385e-02, -1.1467e-02,\n",
       "         -4.2408e-02, -8.2689e-03, -1.5526e-02,  4.1602e-02,  1.9124e-02,\n",
       "          4.7362e-02,  3.4790e-02,  2.5618e-02,  4.9398e-02,  4.0993e-02,\n",
       "          2.6005e-02, -2.3878e-03, -6.2252e-02,  5.4879e-02, -7.3493e-04,\n",
       "         -3.1570e-03,  3.7760e-02, -1.2093e-02,  6.2833e-02, -1.7012e-02,\n",
       "         -3.5599e-02, -8.8798e-03,  4.1763e-02,  1.5609e-02, -5.0999e-02,\n",
       "          1.5573e-02,  4.1263e-02, -7.3341e-02,  2.6701e-02, -3.4302e-02,\n",
       "         -3.4325e-02,  4.4197e-02,  7.9568e-02, -5.4636e-02, -2.1753e-03,\n",
       "          2.5722e-02, -9.0389e-03, -9.0170e-03,  3.6972e-02, -7.3999e-02,\n",
       "          3.0024e-02, -1.7796e-02, -4.7082e-02, -3.7675e-02, -8.3769e-03,\n",
       "         -9.7219e-03,  2.5605e-02,  1.7918e-02,  1.8543e-02,  3.3996e-03,\n",
       "         -4.5072e-02, -1.1612e-03,  1.2849e-04, -5.0495e-03,  1.6710e-02,\n",
       "         -3.3255e-02,  1.1264e-02,  2.4214e-02, -6.5190e-04,  6.1840e-02,\n",
       "         -1.7645e-02,  7.1337e-02,  4.3021e-03, -3.7602e-02,  1.1785e-02,\n",
       "          4.7590e-02, -5.4192e-02, -5.8013e-02, -1.7699e-03, -3.9820e-02,\n",
       "         -2.1892e-02, -9.4714e-03,  3.4002e-02,  8.8190e-03,  4.6538e-02,\n",
       "         -3.4054e-02,  5.7579e-02, -2.7850e-02, -6.2391e-04, -2.6969e-02,\n",
       "         -6.9118e-03,  2.8500e-03,  7.8449e-03,  3.0565e-02, -2.7139e-02,\n",
       "          6.6343e-02, -2.6637e-02, -3.7668e-02, -4.1618e-02,  3.9534e-02,\n",
       "          5.7882e-03,  5.4380e-03, -2.9738e-02,  3.0418e-02, -8.4888e-02,\n",
       "         -4.6304e-02,  3.0901e-02,  4.5770e-02,  4.8970e-02,  5.7202e-02,\n",
       "         -5.2704e-02,  3.2221e-02, -3.0668e-02,  9.4041e-04, -8.0533e-03,\n",
       "          3.2881e-02,  3.6373e-02, -3.7885e-02,  3.8912e-02, -4.4510e-02,\n",
       "          3.2186e-02,  3.9127e-02, -4.2958e-02,  4.2106e-02,  7.5723e-03,\n",
       "         -2.1173e-03, -3.6921e-02,  5.3300e-02,  1.6564e-03,  3.6089e-02,\n",
       "          3.0187e-02,  9.4600e-03, -1.3841e-02, -4.9651e-03,  7.4824e-02,\n",
       "         -1.2495e-02,  2.5470e-03, -9.4435e-02,  5.7172e-02,  1.2908e-02,\n",
       "          2.2644e-02,  1.1458e-02,  3.1269e-02,  1.3064e-02, -1.8486e-02,\n",
       "          7.9458e-03,  3.9575e-02, -1.8308e-02, -2.2467e-02,  2.7360e-02,\n",
       "         -1.0196e-01,  2.9089e-02, -2.5347e-02,  2.7827e-02, -1.1944e-02,\n",
       "          1.9901e-03,  1.0999e-02, -1.7140e-04,  3.2638e-03,  3.3179e-02,\n",
       "         -4.0446e-02, -2.7546e-02, -8.7722e-02, -4.4004e-02, -3.3746e-02,\n",
       "          7.3937e-02, -2.9633e-02, -1.0600e-02, -2.4515e-02, -5.7450e-03,\n",
       "         -1.9652e-02,  1.3041e-02,  1.9995e-02, -1.0792e-02, -9.6748e-03,\n",
       "          1.6739e-02, -3.3757e-02,  5.1222e-02, -1.6436e-02, -1.7407e-02,\n",
       "          4.9783e-02,  2.0186e-02,  4.6983e-02,  2.1397e-02,  2.4617e-02,\n",
       "         -4.0154e-03, -2.4157e-02,  1.1863e-02, -6.6794e-03, -6.7353e-02,\n",
       "         -1.0993e-02, -9.2150e-03,  4.8702e-03, -4.5576e-02, -5.3131e-02,\n",
       "         -3.1957e-02,  1.1374e-02,  3.9534e-02,  3.1755e-02,  3.7509e-02,\n",
       "         -1.0145e-01, -5.1378e-02, -4.8220e-02,  4.7834e-02,  1.9283e-02,\n",
       "          6.9083e-03, -3.7587e-03,  5.0642e-02, -7.3752e-02,  2.4701e-02,\n",
       "          1.4092e-02,  1.4257e-02,  7.2521e-03,  2.6527e-02,  5.1462e-02,\n",
       "          3.1421e-02,  2.1179e-02,  1.3998e-02, -4.4057e-02, -2.6631e-02,\n",
       "          4.5090e-03,  9.2101e-02, -1.3805e-02, -3.4567e-02, -5.4148e-02,\n",
       "         -1.2641e-02, -3.8274e-02,  3.1395e-02,  4.4301e-02,  8.2718e-02,\n",
       "          4.9653e-02,  2.8032e-02, -5.7946e-02, -5.8129e-02, -6.5812e-02,\n",
       "         -3.3387e-03, -2.3723e-02, -9.1257e-03,  4.2736e-02,  3.5852e-02,\n",
       "         -5.3459e-33,  3.2228e-02, -1.7726e-02, -3.4406e-02, -4.8558e-03,\n",
       "         -5.1111e-02,  9.1672e-03, -2.9127e-02,  5.4207e-03,  7.9598e-03,\n",
       "         -2.3982e-02,  2.2571e-02, -5.0037e-02,  7.8441e-03, -1.4766e-02,\n",
       "          3.1875e-02,  1.1801e-02, -1.4402e-02, -5.5198e-02,  2.0529e-02,\n",
       "         -1.5260e-02, -2.0265e-02,  6.1113e-03,  5.1425e-02,  1.1584e-02,\n",
       "         -2.8730e-03, -5.0329e-03, -8.0334e-02,  3.1428e-02, -1.2478e-02,\n",
       "         -1.3210e-03, -2.7684e-02, -1.6095e-02, -3.6808e-03,  3.5659e-03,\n",
       "         -2.0797e-02, -2.1684e-02, -5.0595e-02, -1.1433e-02, -2.7439e-02,\n",
       "          2.1398e-02, -8.5197e-02,  6.8079e-03,  1.0952e-02, -1.2341e-02,\n",
       "         -2.7513e-02, -6.5337e-02, -4.0029e-03, -6.7455e-03, -2.7981e-03,\n",
       "          4.4323e-02,  2.2758e-02,  1.1641e-02,  2.2445e-02,  2.9799e-02,\n",
       "          1.2805e-02, -5.8858e-02, -1.0350e-03,  1.3267e-02,  5.9215e-03,\n",
       "         -4.0886e-02,  5.8834e-02, -4.7097e-02,  9.3731e-03, -6.9022e-03,\n",
       "         -6.1832e-04, -3.5664e-02, -6.1657e-02, -1.3883e-02,  6.1446e-03,\n",
       "          7.1973e-02,  1.1368e-02,  7.4021e-02,  2.1503e-02, -8.8347e-03,\n",
       "          1.6780e-02, -1.1534e-02, -2.3590e-02,  3.8557e-02,  3.9767e-02,\n",
       "          9.9372e-02, -3.1133e-02,  1.6315e-02,  2.2671e-03, -4.5430e-03,\n",
       "         -1.9449e-02, -8.1207e-02, -6.3685e-03, -2.7842e-02,  1.1326e-02,\n",
       "          1.7632e-02,  4.9769e-04,  2.1740e-02, -5.0438e-02,  1.6605e-02,\n",
       "          6.9180e-02,  2.0701e-02,  6.4930e-03, -8.0496e-03, -3.8575e-02,\n",
       "         -4.1740e-02,  3.2981e-02, -4.1215e-02, -2.6449e-03,  2.8814e-02,\n",
       "          3.9445e-02, -2.6006e-02, -8.0692e-03,  1.9256e-02,  2.7365e-02,\n",
       "          1.5098e-02, -7.6030e-03, -1.9737e-02,  1.9772e-02, -6.3508e-03,\n",
       "          9.9343e-03, -2.8411e-02, -1.7064e-02,  7.6845e-02, -4.1481e-02,\n",
       "         -1.0268e-02, -2.0362e-02,  4.2314e-02,  1.5262e-02,  1.2307e-02,\n",
       "         -8.1875e-05, -2.5260e-02,  3.7554e-02,  1.8351e-02,  5.1913e-02,\n",
       "         -5.0176e-02,  2.9041e-02,  2.0803e-02,  2.6475e-07, -3.9805e-02,\n",
       "          5.3842e-02,  6.9280e-03,  2.5193e-02,  2.3088e-02,  4.2227e-02,\n",
       "         -1.1614e-02, -3.6851e-03, -1.5343e-02, -3.7407e-02,  5.2362e-02,\n",
       "          1.9299e-02, -4.1682e-02,  8.0136e-03,  1.3423e-02, -8.0213e-02,\n",
       "          5.4831e-02,  2.6127e-02, -1.3078e-02,  6.9492e-03,  3.5996e-02,\n",
       "          4.3523e-02,  5.9377e-02,  6.0807e-03, -5.3197e-03, -2.5828e-02,\n",
       "         -3.6732e-02, -1.6657e-02, -1.6985e-02, -8.7101e-03, -4.0019e-03,\n",
       "          5.1751e-04, -1.0586e-02,  4.1282e-03, -3.5532e-02, -9.1462e-03,\n",
       "          3.8139e-02,  5.9563e-02, -2.3922e-02,  2.5499e-02,  2.1373e-02,\n",
       "          5.6300e-02, -1.3078e-02, -4.6129e-03, -3.1440e-03, -5.4257e-03,\n",
       "          7.6216e-03, -3.0108e-02, -6.8735e-02,  4.0038e-02, -1.6325e-02,\n",
       "         -8.4979e-03,  4.8839e-02,  4.6409e-02, -1.1969e-04, -6.8069e-03,\n",
       "         -1.7193e-02,  2.6563e-03,  6.5914e-02,  4.6279e-03, -8.4378e-02,\n",
       "          5.3126e-03,  2.2993e-02,  8.9028e-02,  2.3130e-02, -2.2493e-02,\n",
       "         -4.2442e-02,  2.5407e-34, -6.0385e-02, -3.9611e-02,  2.2711e-02,\n",
       "         -3.0863e-03,  1.4210e-02, -7.1206e-03,  3.3401e-02,  1.5450e-02,\n",
       "          2.6818e-02, -8.7164e-02, -5.5824e-02], device='cuda:0')}"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embeddings_text[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "embeddings_pd=pd.DataFrame(embeddings_text)\n",
    "embedding_path=\"embeddings_text.csv\"\n",
    "embeddings_pd.to_csv(embedding_path,index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>page_number</th>\n",
       "      <th>page_char_count</th>\n",
       "      <th>page_token_count</th>\n",
       "      <th>text</th>\n",
       "      <th>embedding</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Neck Flexion', 'equipme...</td>\n",
       "      <td>tensor([ 3.0419e-02, -6.3200e-03, -1.0122e-02,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Neck Flexion', 'equipme...</td>\n",
       "      <td>tensor([ 3.7481e-02, -2.9378e-02, -1.4067e-02,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Lateral Neck Flexion', ...</td>\n",
       "      <td>tensor([ 3.1795e-02, -2.9549e-02, -1.4883e-02,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Neck Flexion', 'equipme...</td>\n",
       "      <td>tensor([ 3.6791e-02, -2.9646e-02, -1.4654e-02,...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>6</td>\n",
       "      <td>1.5</td>\n",
       "      <td>{'exercise': {'name': 'Lateral Neck Flexion', ...</td>\n",
       "      <td>tensor([ 3.2426e-02, -3.0278e-02, -1.5896e-02,...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   page_number  page_char_count  page_token_count  \\\n",
       "0            0                6               1.5   \n",
       "1            1                6               1.5   \n",
       "2            2                6               1.5   \n",
       "3            3                6               1.5   \n",
       "4            4                6               1.5   \n",
       "\n",
       "                                                text  \\\n",
       "0  {'exercise': {'name': 'Neck Flexion', 'equipme...   \n",
       "1  {'exercise': {'name': 'Neck Flexion', 'equipme...   \n",
       "2  {'exercise': {'name': 'Lateral Neck Flexion', ...   \n",
       "3  {'exercise': {'name': 'Neck Flexion', 'equipme...   \n",
       "4  {'exercise': {'name': 'Lateral Neck Flexion', ...   \n",
       "\n",
       "                                           embedding  \n",
       "0  tensor([ 3.0419e-02, -6.3200e-03, -1.0122e-02,...  \n",
       "1  tensor([ 3.7481e-02, -2.9378e-02, -1.4067e-02,...  \n",
       "2  tensor([ 3.1795e-02, -2.9549e-02, -1.4883e-02,...  \n",
       "3  tensor([ 3.6791e-02, -2.9646e-02, -1.4654e-02,...  \n",
       "4  tensor([ 3.2426e-02, -3.0278e-02, -1.5896e-02,...  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df=pd.read_csv(\"embeddings_text.csv\")\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/tmp/ipykernel_2004/1365835288.py:16: DeprecationWarning: string or file could not be read to its end due to unmatched data; this will raise a ValueError in the future.\n",
      "  return np.fromstring(tensor_str, sep=', ').tolist()\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "torch.Size([3702, 768])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "import random\n",
    "\n",
    "import torch\n",
    "import numpy as np \n",
    "import pandas as pd\n",
    "\n",
    "\n",
    "# Import texts and embedding df\n",
    "embedding_text = pd.read_csv(\"embeddings_text.csv\")\n",
    "\n",
    "# Convert embedding column back to np.array (it got converted to string when it got saved to CSV)\n",
    "def tensor_to_list(tensor_str):\n",
    "    # Remove 'tensor(' and ')' from the string\n",
    "    tensor_str = tensor_str.strip('tensor([])')\n",
    "    # Convert the remaining string to a list of floats\n",
    "    return np.fromstring(tensor_str, sep=', ').tolist()\n",
    "\n",
    "df=pd.DataFrame(embedding_text)\n",
    "df.head()\n",
    "# # Apply the conversion\n",
    "device\n",
    "embedding_text[\"embedding\"] = embedding_text[\"embedding\"].apply(tensor_to_list)\n",
    "# Convert texts and embedding df to list of dicts\n",
    "pages_and_chunks = embedding_text.to_dict(orient=\"records\")\n",
    "\n",
    "# Convert embeddings to torch tensor and send to device (note: NumPy arrays are float64, torch tensors are float32 by default)\n",
    "embeddings = torch.tensor(np.array(embedding_text[\"embedding\"].tolist()), dtype=torch.float32).to(device)\n",
    "embeddings.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "tensor([[ 0.0304, -0.0063, -0.0101,  ...,  0.0268, -0.0872, -0.0558],\n",
       "        [ 0.0375, -0.0294, -0.0141,  ...,  0.0282, -0.0848, -0.0600],\n",
       "        [ 0.0318, -0.0295, -0.0149,  ...,  0.0302, -0.0839, -0.0586],\n",
       "        ...,\n",
       "        [ 0.0208, -0.0055, -0.0155,  ...,  0.0461, -0.0455, -0.0393],\n",
       "        [ 0.0251, -0.0095, -0.0186,  ...,  0.0412, -0.0404, -0.0372],\n",
       "        [ 0.0224, -0.0111, -0.0199,  ...,  0.0429, -0.0370, -0.0389]],\n",
       "       device='cuda:0')"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "embeddings"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sentence_transformers import util,SentenceTransformer\n",
    "model=SentenceTransformer(model_name_or_path=\"all-mpnet-base-v2\", \n",
    "                                      device=device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Move both tensors to the same device (GPU in this case)\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "\n",
    "# Ensure the query embedding is on the correct device\n",
    "query_embedding = query_embedding.to(device)\n",
    "\n",
    "# Ensure the embeddings are on the correct device\n",
    "embeddings = embeddings.to(device)\n",
    "\n",
    "# Now compute the dot products\n",
    "dot_products = util.dot_score(a=query_embedding, b=embeddings)[0]\n",
    "\n",
    "# Get the top k products\n",
    "top_products = torch.topk(dot_products, k=10)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'torch.Tensor'>\n",
      "tensor([[ 0.0304, -0.0063, -0.0101,  ...,  0.0268, -0.0872, -0.0558],\n",
      "        [ 0.0375, -0.0294, -0.0141,  ...,  0.0282, -0.0848, -0.0600],\n",
      "        [ 0.0318, -0.0295, -0.0149,  ...,  0.0302, -0.0839, -0.0586],\n",
      "        [ 0.0368, -0.0296, -0.0147,  ...,  0.0293, -0.0847, -0.0602],\n",
      "        [ 0.0324, -0.0303, -0.0159,  ...,  0.0319, -0.0831, -0.0591]],\n",
      "       device='cuda:0')\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "torch.Size([3702, 768])"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "print(type(embeddings))\n",
    "print(embeddings[:5])  # Print the first 5 items to inspect\n",
    "embeddings.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define helper function to print wrapped text \n",
    "import textwrap\n",
    "\n",
    "def print_wrapped(text, wrap_length=80):\n",
    "    wrapped_text = textwrap.fill(text, wrap_length)\n",
    "    print(wrapped_text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Query: 'neck exercises'\n",
      "\n",
      "Results:\n",
      "Score: 0.6202\n",
      "Text:\n",
      "{'exercise': {'name': 'Wall Rear Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Hold small or folded cushioned mat\n",
      "behind head. Facing away from wall or column position head and back of shoulders\n",
      "low on wall or column with mat between. Stand with feet far away from wall or\n",
      "column so body is angled back. Position hips and back straight and bend knees\n",
      "just slightly. Place arms to side or hold hands on abdomen.', 'instructions':\n",
      "'Push head back into mat and roll head upward. Arch spine and straighten knees.\n",
      "Hyperextend neck so head is facing up. Return to original position by rolling\n",
      "head down while allowing low back to straighten and knees to bend slightly.\n",
      "Continue down until back of head and shoulders make contact with mat. Repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Splenius,'],\n",
      "'stabilizer_muscles': ['Gluteus Maximus', '']}, 'workout_parameters': {'sets':\n",
      "3, 'reps': 8, 'bmi_for_exercise': 21.4053265}, 'goals': {'target_audience':\n",
      "'muscle gain', 'difficulty_level': 'Intermediate'}, 'additional_info':\n",
      "{'gender': 'female', 'type_of_plan': 'muscle gain'}}\n",
      "Page number: 17\n",
      "\n",
      "\n",
      "Score: 0.6164\n",
      "Text:\n",
      "{'exercise': {'name': 'Wall Rear Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Hold small or folded cushioned mat\n",
      "behind head. Facing away from wall or column position head and back of shoulders\n",
      "low on wall or column with mat between. Stand with feet far away from wall or\n",
      "column so body is angled back. Position hips and back straight and bend knees\n",
      "just slightly. Place arms to side or hold hands on abdomen.', 'instructions':\n",
      "'Push head back into mat and roll head upward. Arch spine and straighten knees.\n",
      "Hyperextend neck so head is facing up. Return to original position by rolling\n",
      "head down while allowing low back to straighten and knees to bend slightly.\n",
      "Continue down until back of head and shoulders make contact with mat. Repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Splenius,'],\n",
      "'stabilizer_muscles': ['Gluteus Maximus', '']}, 'workout_parameters': {'sets':\n",
      "4, 'reps': 11, 'bmi_for_exercise': 22.18052018}, 'goals': {'target_audience':\n",
      "'weight gain', 'difficulty_level': 'Intermediate'}, 'additional_info':\n",
      "{'gender': 'female', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 2485\n",
      "\n",
      "\n",
      "Score: 0.6149\n",
      "Text:\n",
      "{'exercise': {'name': 'Lateral Neck Flexion', 'equipment': 'Suspended', 'type':\n",
      "'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty':\n",
      "2}, 'execution': {'setup': 'Place belt or strap around end of suspension\n",
      "trainer. Hold belt in hand and stand away from suspension trainer so it hangs to\n",
      "one side. Place belt around head above ears and clamp belt snugly around head by\n",
      "tightly gripping ends close to head above ear closest to suspension trainer.\n",
      "Step further away from suspension trainer until it is taut. Lean away at angle\n",
      "against support of belt and suspension trainer by positioning feet closer to\n",
      "side of suspension trainer while keeping head and body straight.',\n",
      "'instructions': 'Lower angle of body by tilting head to side of suspension\n",
      "trainer. Increase angle of body by tilting head away from side of suspension\n",
      "trainer. Repeat and continue in opposite position.'}, 'muscles': {'main_muscle':\n",
      "'Neck', 'target_muscles': ['Sternocleidomastoid,'], 'stabilizer_muscles': ['']},\n",
      "'workout_parameters': {'sets': 4, 'reps': 14, 'bmi_for_exercise': 22.51264072},\n",
      "'goals': {'target_audience': 'muscle gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'muscle gain'}}\n",
      "Page number: 10\n",
      "\n",
      "\n",
      "Score: 0.6129\n",
      "Text:\n",
      "{'exercise': {'name': 'Wall Side Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Place small or folded cushioned mat on\n",
      "column or at end of wall approximately shoulder height. With hands supporting\n",
      "mat, stand with side toward mat. Step far to side so feet are away from wall or\n",
      "column forward. Place side of head on mat with shoulder extending over end of\n",
      "adjacent side of column or wall so body leans to side. Fold hands behind low\n",
      "back or hips.', 'instructions': 'Push side of head into mat and roll onto top of\n",
      "head moving body away from mat. Return to original position by rolling back down\n",
      "to side of head while bringing shoulder back over end of column or wall.\n",
      "Repeat.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['Gluteus Medius', 'Gluteus\n",
      "Minimus', '']}, 'workout_parameters': {'sets': 4, 'reps': 10,\n",
      "'bmi_for_exercise': 20.84352209}, 'goals': {'target_audience': 'muscle gain',\n",
      "'difficulty_level': 'Intermediate'}, 'additional_info': {'gender': 'female',\n",
      "'type_of_plan': 'muscle gain'}}\n",
      "Page number: 8\n",
      "\n",
      "\n",
      "Score: 0.6125\n",
      "Text:\n",
      "{'exercise': {'name': 'Wall Rear Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Hold small or folded cushioned mat\n",
      "behind head. Facing away from wall or column position head and back of shoulders\n",
      "low on wall or column with mat between. Stand with feet far away from wall or\n",
      "column so body is angled back. Position hips and back straight and bend knees\n",
      "just slightly. Place arms to side or hold hands on abdomen.', 'instructions':\n",
      "'Push head back into mat and roll head upward. Arch spine and straighten knees.\n",
      "Hyperextend neck so head is facing up. Return to original position by rolling\n",
      "head down while allowing low back to straighten and knees to bend slightly.\n",
      "Continue down until back of head and shoulders make contact with mat. Repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Splenius,'],\n",
      "'stabilizer_muscles': ['Gluteus Maximus', '']}, 'workout_parameters': {'sets':\n",
      "5, 'reps': 6, 'bmi_for_exercise': 22.576148570789403}, 'goals':\n",
      "{'target_audience': 'muscle gain', 'difficulty_level': 'Intermediate'},\n",
      "'additional_info': {'gender': 'male', 'type_of_plan': 'muscle gain'}}\n",
      "Page number: 3102\n",
      "\n",
      "\n",
      "Score: 0.6122\n",
      "Text:\n",
      "{'exercise': {'name': 'Wall Side Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Place small or folded cushioned mat on\n",
      "column or at end of wall approximately shoulder height. With hands supporting\n",
      "mat, stand with side toward mat. Step far to side so feet are away from wall or\n",
      "column forward. Place side of head on mat with shoulder extending over end of\n",
      "adjacent side of column or wall so body leans to side. Fold hands behind low\n",
      "back or hips.', 'instructions': 'Push side of head into mat and roll onto top of\n",
      "head moving body away from mat. Return to original position by rolling back down\n",
      "to side of head while bringing shoulder back over end of column or wall.\n",
      "Repeat.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['Gluteus Medius', 'Gluteus\n",
      "Minimus', '']}, 'workout_parameters': {'sets': 3, 'reps': 14,\n",
      "'bmi_for_exercise': 24.21415733}, 'goals': {'target_audience': 'weight gain',\n",
      "'difficulty_level': 'Intermediate'}, 'additional_info': {'gender': 'female',\n",
      "'type_of_plan': 'weight gain'}}\n",
      "Page number: 2476\n",
      "\n",
      "\n",
      "Score: 0.6113\n",
      "Text:\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Weighted', 'type': 'Basic or\n",
      "Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2},\n",
      "'execution': {'setup': 'Place folded towel on weight plate. Lie supine\n",
      "perpendicular on bench with low back and hips extending off and feet on floor.\n",
      "Place weight and towel on forehead. Place plates on forehead with towel placed\n",
      "in between for comfort.', 'instructions': 'Move head up by flexing neck until\n",
      "chin touches upper chest. Return by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
      "'workout_parameters': {'sets': 3, 'reps': 7, 'bmi_for_exercise':\n",
      "22.648034957276007}, 'goals': {'target_audience': 'muscle gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'male',\n",
      "'type_of_plan': 'muscle gain'}}\n",
      "Page number: 3090\n",
      "\n",
      "\n",
      "Score: 0.6109\n",
      "Text:\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Weighted', 'type': 'Basic or\n",
      "Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2},\n",
      "'execution': {'setup': 'Place folded towel on weight plate. Lie supine\n",
      "perpendicular on bench with low back and hips extending off and feet on floor.\n",
      "Place weight and towel on forehead. Place plates on forehead with towel placed\n",
      "in between for comfort.', 'instructions': 'Move head up by flexing neck until\n",
      "chin touches upper chest. Return by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
      "'workout_parameters': {'sets': 4, 'reps': 12, 'bmi_for_exercise': 24.44995135},\n",
      "'goals': {'target_audience': 'weight gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 2473\n",
      "\n",
      "\n",
      "Score: 0.6106\n",
      "Text:\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Weighted', 'type': 'Basic or\n",
      "Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2},\n",
      "'execution': {'setup': 'Place folded towel on weight plate. Lie supine\n",
      "perpendicular on bench with low back and hips extending off and feet on floor.\n",
      "Place weight and towel on forehead. Place plates on forehead with towel placed\n",
      "in between for comfort.', 'instructions': 'Move head up by flexing neck until\n",
      "chin touches upper chest. Return by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
      "'workout_parameters': {'sets': 3, 'reps': 7, 'bmi_for_exercise':\n",
      "22.648034957276007}, 'goals': {'target_audience': 'weight gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'male',\n",
      "'type_of_plan': 'weight gain'}}\n",
      "Page number: 1856\n",
      "\n",
      "\n",
      "Score: 0.6103\n",
      "Text:\n",
      "{'exercise': {'name': 'Lateral Neck Flexion', 'equipment': 'Weighted', 'type':\n",
      "'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty':\n",
      "2}, 'execution': {'setup': 'Place folded towel on weight plate. Lie on bench on\n",
      "side with knees and hips bent and arm hanging over edge. Position weight and\n",
      "towel on side of upper head. Hold weight on side of head with hand of upper arm.\n",
      "Place hand of lower arm on floor for support.', 'instructions': 'Move head up to\n",
      "side by laterally flexing neck. Lower head to opposite side and repeat. Lie on\n",
      "other side and continue.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['']}, 'workout_parameters':\n",
      "{'sets': 3, 'reps': 13, 'bmi_for_exercise': 21.29919937}, 'goals':\n",
      "{'target_audience': 'weight gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 2474\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(f\"Query: '{query}'\\n\")\n",
    "print(\"Results:\")\n",
    "# Loop through zipped together scores and indicies from torch.topk\n",
    "for score, idx in zip(top_products[0], top_products[1]):\n",
    "    print(f\"Score: {score:.4f}\")\n",
    "    # Print relevant sentence chunk (since the scores are in descending order, the most relevant chunk will be first)\n",
    "    print(\"Text:\")\n",
    "    print_wrapped(pages_and_chunks[idx][\"text\"])\n",
    "    # Print the page number too so we can reference the textbook further (and check the results)\n",
    "    print(f\"Page number: {pages_and_chunks[idx]['page_number']}\")\n",
    "    print(\"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [],
   "source": [
    "def retrieve_relevant_resources(query: str,\n",
    "                                embeddings: torch.tensor,\n",
    "                                model: SentenceTransformer=model,\n",
    "                                n_resources_to_return: int=5,\n",
    "                                print_time: bool=True):\n",
    "    \"\"\"\n",
    "    Embeds a query with model and returns top k scores and indices from embeddings.\n",
    "    \"\"\"\n",
    "\n",
    "    # Embed the query\n",
    "    query_embedding = model.encode(query, \n",
    "                                   convert_to_tensor=True) \n",
    "\n",
    "    # Get dot product scores on embeddings\n",
    "    dot_scores = util.dot_score(query_embedding, embeddings)[0]\n",
    "\n",
    "    \n",
    "    scores, indices = torch.topk(input=dot_scores, \n",
    "                                 k=n_resources_to_return)\n",
    "\n",
    "    return scores, indices\n",
    "\n",
    "def print_top_results_and_scores(query: str,\n",
    "                                 embeddings: torch.tensor,\n",
    "                                 pages_and_chunks: list[dict]=pages_and_chunks,\n",
    "                                 n_resources_to_return: int=30):\n",
    "    \"\"\"\n",
    "    Takes a query, retrieves most relevant resources and prints them out in descending order.\n",
    "\n",
    "    Note: Requires pages_and_chunks to be formatted in a specific way (see above for reference).\n",
    "    \"\"\"\n",
    "    \n",
    "    scores, indices = retrieve_relevant_resources(query=query,\n",
    "                                                  embeddings=embeddings,\n",
    "                                                  n_resources_to_return=n_resources_to_return)\n",
    "    \n",
    "    print(f\"Query: {query}\\n\")\n",
    "    print(\"Results:\")\n",
    "    # Loop through zipped together scores and indicies\n",
    "    for score, index in zip(scores, indices):\n",
    "        print(f\"Score: {score:.4f}\")\n",
    "        # Print relevant sentence chunk (since the scores are in descending order, the most relevant chunk will be first)\n",
    "        print_wrapped(pages_and_chunks[index][\"text\"])\n",
    "        # Print the page number too so we can reference the textbook further and check the results\n",
    "        print(f\"Page number: {pages_and_chunks[index]['page_number']}\")\n",
    "        print(\"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(tensor([0.6047, 0.6037, 0.6032, 0.6025, 0.5996, 0.5990, 0.5781, 0.5778, 0.5763,\n",
       "         0.5730, 0.5725, 0.5708, 0.5705, 0.5692, 0.5685, 0.5681, 0.5681, 0.5679,\n",
       "         0.5673, 0.5660, 0.5659, 0.5659, 0.5654, 0.5652, 0.5649, 0.5644, 0.5643,\n",
       "         0.5633, 0.5631, 0.5626], device='cuda:0'),\n",
       " tensor([ 923, 1540, 3391, 2157,  306, 2774, 1553,  936,  319, 3404, 2787, 2170,\n",
       "          933,  899,  298,  909,  929, 3383, 1550,  892, 1546, 1516, 1509, 1526,\n",
       "         1508,  274,  312, 3397,  891,  900], device='cuda:0'))"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "query = \"chest exercise\"\n",
    "\n",
    "# Get just the scores and indices of top related results\n",
    "scores, indices = retrieve_relevant_resources(query=query,\n",
    "                                              embeddings=embeddings,n_resources_to_return=30)\n",
    "scores, indices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Query: neck exercises\n",
      "\n",
      "Results:\n",
      "Score: 0.6202\n",
      "{'exercise': {'name': 'Wall Rear Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Hold small or folded cushioned mat\n",
      "behind head. Facing away from wall or column position head and back of shoulders\n",
      "low on wall or column with mat between. Stand with feet far away from wall or\n",
      "column so body is angled back. Position hips and back straight and bend knees\n",
      "just slightly. Place arms to side or hold hands on abdomen.', 'instructions':\n",
      "'Push head back into mat and roll head upward. Arch spine and straighten knees.\n",
      "Hyperextend neck so head is facing up. Return to original position by rolling\n",
      "head down while allowing low back to straighten and knees to bend slightly.\n",
      "Continue down until back of head and shoulders make contact with mat. Repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Splenius,'],\n",
      "'stabilizer_muscles': ['Gluteus Maximus', '']}, 'workout_parameters': {'sets':\n",
      "3, 'reps': 8, 'bmi_for_exercise': 21.4053265}, 'goals': {'target_audience':\n",
      "'muscle gain', 'difficulty_level': 'Intermediate'}, 'additional_info':\n",
      "{'gender': 'female', 'type_of_plan': 'muscle gain'}}\n",
      "Page number: 17\n",
      "\n",
      "\n",
      "Score: 0.6164\n",
      "{'exercise': {'name': 'Wall Rear Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Hold small or folded cushioned mat\n",
      "behind head. Facing away from wall or column position head and back of shoulders\n",
      "low on wall or column with mat between. Stand with feet far away from wall or\n",
      "column so body is angled back. Position hips and back straight and bend knees\n",
      "just slightly. Place arms to side or hold hands on abdomen.', 'instructions':\n",
      "'Push head back into mat and roll head upward. Arch spine and straighten knees.\n",
      "Hyperextend neck so head is facing up. Return to original position by rolling\n",
      "head down while allowing low back to straighten and knees to bend slightly.\n",
      "Continue down until back of head and shoulders make contact with mat. Repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Splenius,'],\n",
      "'stabilizer_muscles': ['Gluteus Maximus', '']}, 'workout_parameters': {'sets':\n",
      "4, 'reps': 11, 'bmi_for_exercise': 22.18052018}, 'goals': {'target_audience':\n",
      "'weight gain', 'difficulty_level': 'Intermediate'}, 'additional_info':\n",
      "{'gender': 'female', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 2485\n",
      "\n",
      "\n",
      "Score: 0.6149\n",
      "{'exercise': {'name': 'Lateral Neck Flexion', 'equipment': 'Suspended', 'type':\n",
      "'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty':\n",
      "2}, 'execution': {'setup': 'Place belt or strap around end of suspension\n",
      "trainer. Hold belt in hand and stand away from suspension trainer so it hangs to\n",
      "one side. Place belt around head above ears and clamp belt snugly around head by\n",
      "tightly gripping ends close to head above ear closest to suspension trainer.\n",
      "Step further away from suspension trainer until it is taut. Lean away at angle\n",
      "against support of belt and suspension trainer by positioning feet closer to\n",
      "side of suspension trainer while keeping head and body straight.',\n",
      "'instructions': 'Lower angle of body by tilting head to side of suspension\n",
      "trainer. Increase angle of body by tilting head away from side of suspension\n",
      "trainer. Repeat and continue in opposite position.'}, 'muscles': {'main_muscle':\n",
      "'Neck', 'target_muscles': ['Sternocleidomastoid,'], 'stabilizer_muscles': ['']},\n",
      "'workout_parameters': {'sets': 4, 'reps': 14, 'bmi_for_exercise': 22.51264072},\n",
      "'goals': {'target_audience': 'muscle gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'muscle gain'}}\n",
      "Page number: 10\n",
      "\n",
      "\n",
      "Score: 0.6129\n",
      "{'exercise': {'name': 'Wall Side Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Place small or folded cushioned mat on\n",
      "column or at end of wall approximately shoulder height. With hands supporting\n",
      "mat, stand with side toward mat. Step far to side so feet are away from wall or\n",
      "column forward. Place side of head on mat with shoulder extending over end of\n",
      "adjacent side of column or wall so body leans to side. Fold hands behind low\n",
      "back or hips.', 'instructions': 'Push side of head into mat and roll onto top of\n",
      "head moving body away from mat. Return to original position by rolling back down\n",
      "to side of head while bringing shoulder back over end of column or wall.\n",
      "Repeat.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['Gluteus Medius', 'Gluteus\n",
      "Minimus', '']}, 'workout_parameters': {'sets': 4, 'reps': 10,\n",
      "'bmi_for_exercise': 20.84352209}, 'goals': {'target_audience': 'muscle gain',\n",
      "'difficulty_level': 'Intermediate'}, 'additional_info': {'gender': 'female',\n",
      "'type_of_plan': 'muscle gain'}}\n",
      "Page number: 8\n",
      "\n",
      "\n",
      "Score: 0.6125\n",
      "{'exercise': {'name': 'Wall Rear Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Hold small or folded cushioned mat\n",
      "behind head. Facing away from wall or column position head and back of shoulders\n",
      "low on wall or column with mat between. Stand with feet far away from wall or\n",
      "column so body is angled back. Position hips and back straight and bend knees\n",
      "just slightly. Place arms to side or hold hands on abdomen.', 'instructions':\n",
      "'Push head back into mat and roll head upward. Arch spine and straighten knees.\n",
      "Hyperextend neck so head is facing up. Return to original position by rolling\n",
      "head down while allowing low back to straighten and knees to bend slightly.\n",
      "Continue down until back of head and shoulders make contact with mat. Repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Splenius,'],\n",
      "'stabilizer_muscles': ['Gluteus Maximus', '']}, 'workout_parameters': {'sets':\n",
      "5, 'reps': 6, 'bmi_for_exercise': 22.576148570789403}, 'goals':\n",
      "{'target_audience': 'muscle gain', 'difficulty_level': 'Intermediate'},\n",
      "'additional_info': {'gender': 'male', 'type_of_plan': 'muscle gain'}}\n",
      "Page number: 3102\n",
      "\n",
      "\n",
      "Score: 0.6122\n",
      "{'exercise': {'name': 'Wall Side Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Place small or folded cushioned mat on\n",
      "column or at end of wall approximately shoulder height. With hands supporting\n",
      "mat, stand with side toward mat. Step far to side so feet are away from wall or\n",
      "column forward. Place side of head on mat with shoulder extending over end of\n",
      "adjacent side of column or wall so body leans to side. Fold hands behind low\n",
      "back or hips.', 'instructions': 'Push side of head into mat and roll onto top of\n",
      "head moving body away from mat. Return to original position by rolling back down\n",
      "to side of head while bringing shoulder back over end of column or wall.\n",
      "Repeat.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['Gluteus Medius', 'Gluteus\n",
      "Minimus', '']}, 'workout_parameters': {'sets': 3, 'reps': 14,\n",
      "'bmi_for_exercise': 24.21415733}, 'goals': {'target_audience': 'weight gain',\n",
      "'difficulty_level': 'Intermediate'}, 'additional_info': {'gender': 'female',\n",
      "'type_of_plan': 'weight gain'}}\n",
      "Page number: 2476\n",
      "\n",
      "\n",
      "Score: 0.6113\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Weighted', 'type': 'Basic or\n",
      "Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2},\n",
      "'execution': {'setup': 'Place folded towel on weight plate. Lie supine\n",
      "perpendicular on bench with low back and hips extending off and feet on floor.\n",
      "Place weight and towel on forehead. Place plates on forehead with towel placed\n",
      "in between for comfort.', 'instructions': 'Move head up by flexing neck until\n",
      "chin touches upper chest. Return by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
      "'workout_parameters': {'sets': 3, 'reps': 7, 'bmi_for_exercise':\n",
      "22.648034957276007}, 'goals': {'target_audience': 'muscle gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'male',\n",
      "'type_of_plan': 'muscle gain'}}\n",
      "Page number: 3090\n",
      "\n",
      "\n",
      "Score: 0.6109\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Weighted', 'type': 'Basic or\n",
      "Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2},\n",
      "'execution': {'setup': 'Place folded towel on weight plate. Lie supine\n",
      "perpendicular on bench with low back and hips extending off and feet on floor.\n",
      "Place weight and towel on forehead. Place plates on forehead with towel placed\n",
      "in between for comfort.', 'instructions': 'Move head up by flexing neck until\n",
      "chin touches upper chest. Return by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
      "'workout_parameters': {'sets': 4, 'reps': 12, 'bmi_for_exercise': 24.44995135},\n",
      "'goals': {'target_audience': 'weight gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 2473\n",
      "\n",
      "\n",
      "Score: 0.6106\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Weighted', 'type': 'Basic or\n",
      "Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2},\n",
      "'execution': {'setup': 'Place folded towel on weight plate. Lie supine\n",
      "perpendicular on bench with low back and hips extending off and feet on floor.\n",
      "Place weight and towel on forehead. Place plates on forehead with towel placed\n",
      "in between for comfort.', 'instructions': 'Move head up by flexing neck until\n",
      "chin touches upper chest. Return by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
      "'workout_parameters': {'sets': 3, 'reps': 7, 'bmi_for_exercise':\n",
      "22.648034957276007}, 'goals': {'target_audience': 'weight gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'male',\n",
      "'type_of_plan': 'weight gain'}}\n",
      "Page number: 1856\n",
      "\n",
      "\n",
      "Score: 0.6103\n",
      "{'exercise': {'name': 'Lateral Neck Flexion', 'equipment': 'Weighted', 'type':\n",
      "'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty':\n",
      "2}, 'execution': {'setup': 'Place folded towel on weight plate. Lie on bench on\n",
      "side with knees and hips bent and arm hanging over edge. Position weight and\n",
      "towel on side of upper head. Hold weight on side of head with hand of upper arm.\n",
      "Place hand of lower arm on floor for support.', 'instructions': 'Move head up to\n",
      "side by laterally flexing neck. Lower head to opposite side and repeat. Lie on\n",
      "other side and continue.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['']}, 'workout_parameters':\n",
      "{'sets': 3, 'reps': 13, 'bmi_for_exercise': 21.29919937}, 'goals':\n",
      "{'target_audience': 'weight gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 2474\n",
      "\n",
      "\n",
      "Score: 0.6101\n",
      "{'exercise': {'name': 'Wall Side Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Place small or folded cushioned mat on\n",
      "column or at end of wall approximately shoulder height. With hands supporting\n",
      "mat, stand with side toward mat. Step far to side so feet are away from wall or\n",
      "column forward. Place side of head on mat with shoulder extending over end of\n",
      "adjacent side of column or wall so body leans to side. Fold hands behind low\n",
      "back or hips.', 'instructions': 'Push side of head into mat and roll onto top of\n",
      "head moving body away from mat. Return to original position by rolling back down\n",
      "to side of head while bringing shoulder back over end of column or wall.\n",
      "Repeat.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['Gluteus Medius', 'Gluteus\n",
      "Minimus', '']}, 'workout_parameters': {'sets': 5, 'reps': 10,\n",
      "'bmi_for_exercise': 23.763676966196023}, 'goals': {'target_audience': 'muscle\n",
      "gain', 'difficulty_level': 'Intermediate'}, 'additional_info': {'gender':\n",
      "'male', 'type_of_plan': 'muscle gain'}}\n",
      "Page number: 3093\n",
      "\n",
      "\n",
      "Score: 0.6100\n",
      "{'exercise': {'name': 'Lateral Neck Flexion', 'equipment': 'Weighted', 'type':\n",
      "'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty':\n",
      "2}, 'execution': {'setup': 'Place folded towel on weight plate. Lie on bench on\n",
      "side with knees and hips bent and arm hanging over edge. Position weight and\n",
      "towel on side of upper head. Hold weight on side of head with hand of upper arm.\n",
      "Place hand of lower arm on floor for support.', 'instructions': 'Move head up to\n",
      "side by laterally flexing neck. Lower head to opposite side and repeat. Lie on\n",
      "other side and continue.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['']}, 'workout_parameters':\n",
      "{'sets': 5, 'reps': 10, 'bmi_for_exercise': 21.24120759}, 'goals':\n",
      "{'target_audience': 'muscle gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'muscle gain'}}\n",
      "Page number: 6\n",
      "\n",
      "\n",
      "Score: 0.6097\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Weighted', 'type': 'Basic or\n",
      "Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2},\n",
      "'execution': {'setup': 'Place folded towel on weight plate. Lie supine\n",
      "perpendicular on bench with low back and hips extending off and feet on floor.\n",
      "Place weight and towel on forehead. Place plates on forehead with towel placed\n",
      "in between for comfort.', 'instructions': 'Move head up by flexing neck until\n",
      "chin touches upper chest. Return by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
      "'workout_parameters': {'sets': 4, 'reps': 12, 'bmi_for_exercise': 20.82859637},\n",
      "'goals': {'target_audience': 'muscle gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'muscle gain'}}\n",
      "Page number: 5\n",
      "\n",
      "\n",
      "Score: 0.6093\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Weighted', 'type': 'Basic or\n",
      "Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2},\n",
      "'execution': {'setup': 'Place folded towel on weight plate. Lie supine\n",
      "perpendicular on bench with low back and hips extending off and feet on floor.\n",
      "Place weight and towel on forehead. Place plates on forehead with towel placed\n",
      "in between for comfort.', 'instructions': 'Move head up by flexing neck until\n",
      "chin touches upper chest. Return by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
      "'workout_parameters': {'sets': 3, 'reps': 13, 'bmi_for_exercise': 22.4},\n",
      "'goals': {'target_audience': 'weight loss', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'weight loss'}}\n",
      "Page number: 622\n",
      "\n",
      "\n",
      "Score: 0.6089\n",
      "{'exercise': {'name': 'Lateral Neck Flexion', 'equipment': 'Suspended', 'type':\n",
      "'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty':\n",
      "2}, 'execution': {'setup': 'Place belt or strap around end of suspension\n",
      "trainer. Hold belt in hand and stand away from suspension trainer so it hangs to\n",
      "one side. Place belt around head above ears and clamp belt snugly around head by\n",
      "tightly gripping ends close to head above ear closest to suspension trainer.\n",
      "Step further away from suspension trainer until it is taut. Lean away at angle\n",
      "against support of belt and suspension trainer by positioning feet closer to\n",
      "side of suspension trainer while keeping head and body straight.',\n",
      "'instructions': 'Lower angle of body by tilting head to side of suspension\n",
      "trainer. Increase angle of body by tilting head away from side of suspension\n",
      "trainer. Repeat and continue in opposite position.'}, 'muscles': {'main_muscle':\n",
      "'Neck', 'target_muscles': ['Sternocleidomastoid,'], 'stabilizer_muscles': ['']},\n",
      "'workout_parameters': {'sets': 4, 'reps': 10, 'bmi_for_exercise':\n",
      "22.52473993885878}, 'goals': {'target_audience': 'muscle gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'male',\n",
      "'type_of_plan': 'muscle gain'}}\n",
      "Page number: 3095\n",
      "\n",
      "\n",
      "Score: 0.6079\n",
      "{'exercise': {'name': 'Lateral Neck Flexion', 'equipment': 'Suspended', 'type':\n",
      "'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty':\n",
      "2}, 'execution': {'setup': 'Place belt or strap around end of suspension\n",
      "trainer. Hold belt in hand and stand away from suspension trainer so it hangs to\n",
      "one side. Place belt around head above ears and clamp belt snugly around head by\n",
      "tightly gripping ends close to head above ear closest to suspension trainer.\n",
      "Step further away from suspension trainer until it is taut. Lean away at angle\n",
      "against support of belt and suspension trainer by positioning feet closer to\n",
      "side of suspension trainer while keeping head and body straight.',\n",
      "'instructions': 'Lower angle of body by tilting head to side of suspension\n",
      "trainer. Increase angle of body by tilting head away from side of suspension\n",
      "trainer. Repeat and continue in opposite position.'}, 'muscles': {'main_muscle':\n",
      "'Neck', 'target_muscles': ['Sternocleidomastoid,'], 'stabilizer_muscles': ['']},\n",
      "'workout_parameters': {'sets': 4, 'reps': 14, 'bmi_for_exercise': 22.01666177},\n",
      "'goals': {'target_audience': 'weight gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 2478\n",
      "\n",
      "\n",
      "Score: 0.6077\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Weighted', 'type': 'Basic or\n",
      "Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty': 2},\n",
      "'execution': {'setup': 'Place folded towel on weight plate. Lie supine\n",
      "perpendicular on bench with low back and hips extending off and feet on floor.\n",
      "Place weight and towel on forehead. Place plates on forehead with towel placed\n",
      "in between for comfort.', 'instructions': 'Move head up by flexing neck until\n",
      "chin touches upper chest. Return by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Rectus Abdominis', 'Obliques', '']},\n",
      "'workout_parameters': {'sets': 3, 'reps': 13, 'bmi_for_exercise': 22.6},\n",
      "'goals': {'target_audience': 'weight loss', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'male', 'type_of_plan': 'weight loss'}}\n",
      "Page number: 1239\n",
      "\n",
      "\n",
      "Score: 0.6068\n",
      "{'exercise': {'name': 'Wall Rear Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Hold small or folded cushioned mat\n",
      "behind head. Facing away from wall or column position head and back of shoulders\n",
      "low on wall or column with mat between. Stand with feet far away from wall or\n",
      "column so body is angled back. Position hips and back straight and bend knees\n",
      "just slightly. Place arms to side or hold hands on abdomen.', 'instructions':\n",
      "'Push head back into mat and roll head upward. Arch spine and straighten knees.\n",
      "Hyperextend neck so head is facing up. Return to original position by rolling\n",
      "head down while allowing low back to straighten and knees to bend slightly.\n",
      "Continue down until back of head and shoulders make contact with mat. Repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Splenius,'],\n",
      "'stabilizer_muscles': ['Gluteus Maximus', '']}, 'workout_parameters': {'sets':\n",
      "5, 'reps': 6, 'bmi_for_exercise': 22.576148570789403}, 'goals':\n",
      "{'target_audience': 'weight gain', 'difficulty_level': 'Intermediate'},\n",
      "'additional_info': {'gender': 'male', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 1868\n",
      "\n",
      "\n",
      "Score: 0.6067\n",
      "{'exercise': {'name': 'Wall Rear Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Hold small or folded cushioned mat\n",
      "behind head. Facing away from wall or column position head and back of shoulders\n",
      "low on wall or column with mat between. Stand with feet far away from wall or\n",
      "column so body is angled back. Position hips and back straight and bend knees\n",
      "just slightly. Place arms to side or hold hands on abdomen.', 'instructions':\n",
      "'Push head back into mat and roll head upward. Arch spine and straighten knees.\n",
      "Hyperextend neck so head is facing up. Return to original position by rolling\n",
      "head down while allowing low back to straighten and knees to bend slightly.\n",
      "Continue down until back of head and shoulders make contact with mat. Repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Splenius,'],\n",
      "'stabilizer_muscles': ['Gluteus Maximus', '']}, 'workout_parameters': {'sets':\n",
      "5, 'reps': 11, 'bmi_for_exercise': 23.1}, 'goals': {'target_audience': 'weight\n",
      "loss', 'difficulty_level': 'Intermediate'}, 'additional_info': {'gender':\n",
      "'male', 'type_of_plan': 'weight loss'}}\n",
      "Page number: 1251\n",
      "\n",
      "\n",
      "Score: 0.6066\n",
      "{'exercise': {'name': 'Wall Rear Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Hold small or folded cushioned mat\n",
      "behind head. Facing away from wall or column position head and back of shoulders\n",
      "low on wall or column with mat between. Stand with feet far away from wall or\n",
      "column so body is angled back. Position hips and back straight and bend knees\n",
      "just slightly. Place arms to side or hold hands on abdomen.', 'instructions':\n",
      "'Push head back into mat and roll head upward. Arch spine and straighten knees.\n",
      "Hyperextend neck so head is facing up. Return to original position by rolling\n",
      "head down while allowing low back to straighten and knees to bend slightly.\n",
      "Continue down until back of head and shoulders make contact with mat. Repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Splenius,'],\n",
      "'stabilizer_muscles': ['Gluteus Maximus', '']}, 'workout_parameters': {'sets':\n",
      "4, 'reps': 13, 'bmi_for_exercise': 21.7}, 'goals': {'target_audience': 'weight\n",
      "loss', 'difficulty_level': 'Intermediate'}, 'additional_info': {'gender':\n",
      "'female', 'type_of_plan': 'weight loss'}}\n",
      "Page number: 634\n",
      "\n",
      "\n",
      "Score: 0.6066\n",
      "{'exercise': {'name': 'Wall Side Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Place small or folded cushioned mat on\n",
      "column or at end of wall approximately shoulder height. With hands supporting\n",
      "mat, stand with side toward mat. Step far to side so feet are away from wall or\n",
      "column forward. Place side of head on mat with shoulder extending over end of\n",
      "adjacent side of column or wall so body leans to side. Fold hands behind low\n",
      "back or hips.', 'instructions': 'Push side of head into mat and roll onto top of\n",
      "head moving body away from mat. Return to original position by rolling back down\n",
      "to side of head while bringing shoulder back over end of column or wall.\n",
      "Repeat.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['Gluteus Medius', 'Gluteus\n",
      "Minimus', '']}, 'workout_parameters': {'sets': 5, 'reps': 10,\n",
      "'bmi_for_exercise': 23.763676966196023}, 'goals': {'target_audience': 'weight\n",
      "gain', 'difficulty_level': 'Intermediate'}, 'additional_info': {'gender':\n",
      "'male', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 1859\n",
      "\n",
      "\n",
      "Score: 0.6060\n",
      "{'exercise': {'name': 'Seated Neck Extension', 'equipment': 'Weighted', 'type':\n",
      "'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 2}, 'execution': {'setup': 'Place folded towel on weight plate. On\n",
      "bench or stool, sit bent over on thighs. With both hands, place plate(s) on back\n",
      "of head with towel placed in between for comfort.', 'instructions': 'Move head\n",
      "up by hyperextending neck up as high as possible. Return by bending neck down\n",
      "until chin touches upper chest. Repeat.'}, 'muscles': {'main_muscle': 'Neck',\n",
      "'target_muscles': ['Splenius,'], 'stabilizer_muscles': ['No significant\n",
      "stabilizers', '']}, 'workout_parameters': {'sets': 4, 'reps': 12,\n",
      "'bmi_for_exercise': 23.18671677}, 'goals': {'target_audience': 'weight gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'female',\n",
      "'type_of_plan': 'weight gain'}}\n",
      "Page number: 2482\n",
      "\n",
      "\n",
      "Score: 0.6056\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Lever (plate loaded)',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 2}, 'execution': {'setup': 'Sit on seat in machine. Position\n",
      "padded lever on face. Position body so neck is in line with lever axis. Grasp\n",
      "handles for support.', 'instructions': 'Move head forward by flexing neck until\n",
      "chin touches upper chest. Return head by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Latissimus Dorsi', 'Deltoid', 'Posterior', 'Rhomboids',\n",
      "'Trapezius', 'Middle', 'Trapezius', 'Lower', 'Triceps Brachii', 'Rectus\n",
      "Abdominis', 'Obliques', '']}, 'workout_parameters': {'sets': 3, 'reps': 12,\n",
      "'bmi_for_exercise': 22.29599736}, 'goals': {'target_audience': 'weight gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'female',\n",
      "'type_of_plan': 'weight gain'}}\n",
      "Page number: 2469\n",
      "\n",
      "\n",
      "Score: 0.6055\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Lever (selectorized)',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 2}, 'execution': {'setup': 'Sit on seat in machine. Position\n",
      "padded lever on face. Position body so neck is in line with lever axis. Grasp\n",
      "handles for support.', 'instructions': 'Move head forward by flexing neck until\n",
      "chin touches upper chest. Return head by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Latissimus Dorsi', 'Deltoid', 'Posterior', 'Rhomboids',\n",
      "'Trapezius', 'Middle', 'Trapezius', 'Lower', 'Triceps Brachii', 'Rectus\n",
      "Abdominis', 'Obliques', '']}, 'workout_parameters': {'sets': 3, 'reps': 12,\n",
      "'bmi_for_exercise': 21.18925321}, 'goals': {'target_audience': 'weight gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'female',\n",
      "'type_of_plan': 'weight gain'}}\n",
      "Page number: 2471\n",
      "\n",
      "\n",
      "Score: 0.6049\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Lever (plate loaded)',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 2}, 'execution': {'setup': 'Sit on seat in machine. Position\n",
      "padded lever on face. Position body so neck is in line with lever axis. Grasp\n",
      "handles for support.', 'instructions': 'Move head forward by flexing neck until\n",
      "chin touches upper chest. Return head by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Latissimus Dorsi', 'Deltoid', 'Posterior', 'Rhomboids',\n",
      "'Trapezius', 'Middle', 'Trapezius', 'Lower', 'Triceps Brachii', 'Rectus\n",
      "Abdominis', 'Obliques', '']}, 'workout_parameters': {'sets': 4, 'reps': 6,\n",
      "'bmi_for_exercise': 22.314584232919188}, 'goals': {'target_audience': 'muscle\n",
      "gain', 'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'male',\n",
      "'type_of_plan': 'muscle gain'}}\n",
      "Page number: 3086\n",
      "\n",
      "\n",
      "Score: 0.6045\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Lever (plate loaded)',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 2}, 'execution': {'setup': 'Sit on seat in machine. Position\n",
      "padded lever on face. Position body so neck is in line with lever axis. Grasp\n",
      "handles for support.', 'instructions': 'Move head forward by flexing neck until\n",
      "chin touches upper chest. Return head by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Latissimus Dorsi', 'Deltoid', 'Posterior', 'Rhomboids',\n",
      "'Trapezius', 'Middle', 'Trapezius', 'Lower', 'Triceps Brachii', 'Rectus\n",
      "Abdominis', 'Obliques', '']}, 'workout_parameters': {'sets': 4, 'reps': 6,\n",
      "'bmi_for_exercise': 22.314584232919188}, 'goals': {'target_audience': 'weight\n",
      "gain', 'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'male',\n",
      "'type_of_plan': 'weight gain'}}\n",
      "Page number: 1852\n",
      "\n",
      "\n",
      "Score: 0.6044\n",
      "{'exercise': {'name': 'Neck Flexion', 'equipment': 'Lever (plate loaded)',\n",
      "'type': 'Basic or Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 2}, 'execution': {'setup': 'Sit on seat in machine. Position\n",
      "padded lever on face. Position body so neck is in line with lever axis. Grasp\n",
      "handles for support.', 'instructions': 'Move head forward by flexing neck until\n",
      "chin touches upper chest. Return head by hyperextending neck and repeat.'},\n",
      "'muscles': {'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Latissimus Dorsi', 'Deltoid', 'Posterior', 'Rhomboids',\n",
      "'Trapezius', 'Middle', 'Trapezius', 'Lower', 'Triceps Brachii', 'Rectus\n",
      "Abdominis', 'Obliques', '']}, 'workout_parameters': {'sets': 3, 'reps': 14,\n",
      "'bmi_for_exercise': 21.00525585}, 'goals': {'target_audience': 'muscle gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'female',\n",
      "'type_of_plan': 'muscle gain'}}\n",
      "Page number: 1\n",
      "\n",
      "\n",
      "Score: 0.6040\n",
      "{'exercise': {'name': 'Wall Side Neck Bridge', 'equipment': 'Body Weight',\n",
      "'type': 'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull',\n",
      "'difficulty': 3}, 'execution': {'setup': 'Place small or folded cushioned mat on\n",
      "column or at end of wall approximately shoulder height. With hands supporting\n",
      "mat, stand with side toward mat. Step far to side so feet are away from wall or\n",
      "column forward. Place side of head on mat with shoulder extending over end of\n",
      "adjacent side of column or wall so body leans to side. Fold hands behind low\n",
      "back or hips.', 'instructions': 'Push side of head into mat and roll onto top of\n",
      "head moving body away from mat. Return to original position by rolling back down\n",
      "to side of head while bringing shoulder back over end of column or wall.\n",
      "Repeat.'}, 'muscles': {'main_muscle': 'Neck', 'target_muscles':\n",
      "['Sternocleidomastoid,'], 'stabilizer_muscles': ['Gluteus Medius', 'Gluteus\n",
      "Minimus', '']}, 'workout_parameters': {'sets': 3, 'reps': 12,\n",
      "'bmi_for_exercise': 23.1}, 'goals': {'target_audience': 'weight loss',\n",
      "'difficulty_level': 'Intermediate'}, 'additional_info': {'gender': 'male',\n",
      "'type_of_plan': 'weight loss'}}\n",
      "Page number: 1242\n",
      "\n",
      "\n",
      "Score: 0.6039\n",
      "{'exercise': {'name': 'Lateral Neck Flexion', 'equipment': 'Lever (plate\n",
      "loaded)', 'type': 'Auxiliary', 'movement_type': 'Isolated', 'force_type':\n",
      "'Pull', 'difficulty': 2}, 'execution': {'setup': 'Sit on seat in machine with\n",
      "feet apart   . Position padded lever on side of head. Center body so neck is in\n",
      "line with lever axis. Grasp handles for support.', 'instructions': 'Move head\n",
      "down to side by laterally flexing neck. Return head to opposite side and repeat.\n",
      "Sit in machine backwards and continue with opposite side.'}, 'muscles':\n",
      "{'main_muscle': 'Neck', 'target_muscles': ['Sternocleidomastoid,'],\n",
      "'stabilizer_muscles': ['Latissimus Dorsi', 'Pectoralis Major', 'Sternal',\n",
      "'Pectoralis Major', 'Clavicular', 'Coracobrachialis', 'Pectoralis Minor', '']},\n",
      "'workout_parameters': {'sets': 3, 'reps': 11, 'bmi_for_exercise': 23.890636},\n",
      "'goals': {'target_audience': 'weight gain', 'difficulty_level': 'Beginner'},\n",
      "'additional_info': {'gender': 'female', 'type_of_plan': 'weight gain'}}\n",
      "Page number: 2470\n",
      "\n",
      "\n",
      "Score: 0.6036\n",
      "{'exercise': {'name': 'Lateral Neck Flexion', 'equipment': 'Suspended', 'type':\n",
      "'Auxiliary', 'movement_type': 'Isolated', 'force_type': 'Pull', 'difficulty':\n",
      "2}, 'execution': {'setup': 'Place belt or strap around end of suspension\n",
      "trainer. Hold belt in hand and stand away from suspension trainer so it hangs to\n",
      "one side. Place belt around head above ears and clamp belt snugly around head by\n",
      "tightly gripping ends close to head above ear closest to suspension trainer.\n",
      "Step further away from suspension trainer until it is taut. Lean away at angle\n",
      "against support of belt and suspension trainer by positioning feet closer to\n",
      "side of suspension trainer while keeping head and body straight.',\n",
      "'instructions': 'Lower angle of body by tilting head to side of suspension\n",
      "trainer. Increase angle of body by tilting head away from side of suspension\n",
      "trainer. Repeat and continue in opposite position.'}, 'muscles': {'main_muscle':\n",
      "'Neck', 'target_muscles': ['Sternocleidomastoid,'], 'stabilizer_muscles': ['']},\n",
      "'workout_parameters': {'sets': 4, 'reps': 10, 'bmi_for_exercise':\n",
      "22.52473993885878}, 'goals': {'target_audience': 'weight gain',\n",
      "'difficulty_level': 'Beginner'}, 'additional_info': {'gender': 'male',\n",
      "'type_of_plan': 'weight gain'}}\n",
      "Page number: 1861\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print_top_results_and_scores(query=query,\n",
    "                             embeddings=embeddings)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Available GPU memory: 15 GB\n"
     ]
    }
   ],
   "source": [
    "# Get GPU available memory\n",
    "import torch\n",
    "gpu_memory_bytes = torch.cuda.get_device_properties(0).total_memory\n",
    "gpu_memory_gb = round(gpu_memory_bytes / (2**30))\n",
    "print(f\"Available GPU memory: {gpu_memory_gb} GB\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "GPU memory: 15 | Recommended model: Gemma 2B in float16 or Gemma 7B in 4-bit precision.\n",
      "use_quantization_config set to: False\n",
      "model_id set to: google/gemma-2b-it\n"
     ]
    }
   ],
   "source": [
    "# Note: the following is Gemma focused, however, there are more and more LLMs of the 2B and 7B size appearing for local use.\n",
    "if gpu_memory_gb < 5.1:\n",
    "    print(f\"Your available GPU memory is {gpu_memory_gb}GB, you may not have enough memory to run a Gemma LLM locally without quantization.\")\n",
    "elif gpu_memory_gb < 8.1:\n",
    "    print(f\"GPU memory: {gpu_memory_gb} | Recommended model: Gemma 2B in 4-bit precision.\")\n",
    "    use_quantization_config = True \n",
    "    model_id = \"google/gemma-2b-it\"\n",
    "elif gpu_memory_gb < 19.0:\n",
    "    print(f\"GPU memory: {gpu_memory_gb} | Recommended model: Gemma 2B in float16 or Gemma 7B in 4-bit precision.\")\n",
    "    use_quantization_config = False \n",
    "    model_id = \"google/gemma-2b-it\"\n",
    "elif gpu_memory_gb > 19.0:\n",
    "    print(f\"GPU memory: {gpu_memory_gb} | Recommend model: Gemma 7B in 4-bit or float16 precision.\")\n",
    "    use_quantization_config = False \n",
    "    model_id = \"google/gemma-7b-it\"\n",
    "\n",
    "print(f\"use_quantization_config set to: {use_quantization_config}\")\n",
    "print(f\"model_id set to: {model_id}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
      "To disable this warning, you can either:\n",
      "\t- Avoid using `tokenizers` before the fork if possible\n",
      "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: bitsandbytes in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (0.43.1)\n",
      "Requirement already satisfied: accelerate in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (0.32.1)\n",
      "Requirement already satisfied: torch in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from bitsandbytes) (2.2.1+cu121)\n",
      "Requirement already satisfied: numpy in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from bitsandbytes) (1.26.4)\n",
      "Requirement already satisfied: packaging>=20.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from accelerate) (24.1)\n",
      "Requirement already satisfied: psutil in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from accelerate) (6.0.0)\n",
      "Requirement already satisfied: pyyaml in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from accelerate) (6.0.1)\n",
      "Requirement already satisfied: huggingface-hub in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from accelerate) (0.23.4)\n",
      "Requirement already satisfied: safetensors>=0.3.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from accelerate) (0.4.3)\n",
      "Requirement already satisfied: filelock in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (3.15.4)\n",
      "Requirement already satisfied: typing-extensions>=4.8.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (4.12.2)\n",
      "Requirement already satisfied: sympy in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (1.12.1)\n",
      "Requirement already satisfied: networkx in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (3.3)\n",
      "Requirement already satisfied: jinja2 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (3.1.4)\n",
      "Requirement already satisfied: fsspec in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (2024.6.1)\n",
      "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cudnn-cu12==8.9.2.26 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (8.9.2.26)\n",
      "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (12.1.3.1)\n",
      "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (11.0.2.54)\n",
      "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (10.3.2.106)\n",
      "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (11.4.5.107)\n",
      "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (12.1.0.106)\n",
      "Requirement already satisfied: nvidia-nccl-cu12==2.19.3 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (2.19.3)\n",
      "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (12.1.105)\n",
      "Requirement already satisfied: triton==2.2.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->bitsandbytes) (2.2.0)\n",
      "Requirement already satisfied: nvidia-nvjitlink-cu12 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch->bitsandbytes) (12.5.82)\n",
      "Requirement already satisfied: requests in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub->accelerate) (2.32.3)\n",
      "Requirement already satisfied: tqdm>=4.42.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from huggingface-hub->accelerate) (4.66.4)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from jinja2->torch->bitsandbytes) (2.1.5)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests->huggingface-hub->accelerate) (3.3.2)\n",
      "Requirement already satisfied: idna<4,>=2.5 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests->huggingface-hub->accelerate) (3.7)\n",
      "Requirement already satisfied: urllib3<3,>=1.21.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests->huggingface-hub->accelerate) (2.2.2)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from requests->huggingface-hub->accelerate) (2024.7.4)\n",
      "Requirement already satisfied: mpmath<1.4.0,>=1.1.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sympy->torch->bitsandbytes) (1.3.0)\n",
      "\n",
      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.1.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.2\u001b[0m\n",
      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\n",
      "To disable this warning, you can either:\n",
      "\t- Avoid using `tokenizers` before the fork if possible\n",
      "\t- Explicitly set the environment variable TOKENIZERS_PARALLELISM=(true | false)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: flash-attn in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (2.5.9.post1)\n",
      "Requirement already satisfied: torch in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from flash-attn) (2.2.1+cu121)\n",
      "Requirement already satisfied: einops in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from flash-attn) (0.8.0)\n",
      "Requirement already satisfied: filelock in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (3.15.4)\n",
      "Requirement already satisfied: typing-extensions>=4.8.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (4.12.2)\n",
      "Requirement already satisfied: sympy in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (1.12.1)\n",
      "Requirement already satisfied: networkx in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (3.3)\n",
      "Requirement already satisfied: jinja2 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (3.1.4)\n",
      "Requirement already satisfied: fsspec in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (2024.6.1)\n",
      "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (12.1.105)\n",
      "Requirement already satisfied: nvidia-cudnn-cu12==8.9.2.26 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (8.9.2.26)\n",
      "Requirement already satisfied: nvidia-cublas-cu12==12.1.3.1 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (12.1.3.1)\n",
      "Requirement already satisfied: nvidia-cufft-cu12==11.0.2.54 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (11.0.2.54)\n",
      "Requirement already satisfied: nvidia-curand-cu12==10.3.2.106 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (10.3.2.106)\n",
      "Requirement already satisfied: nvidia-cusolver-cu12==11.4.5.107 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (11.4.5.107)\n",
      "Requirement already satisfied: nvidia-cusparse-cu12==12.1.0.106 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (12.1.0.106)\n",
      "Requirement already satisfied: nvidia-nccl-cu12==2.19.3 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (2.19.3)\n",
      "Requirement already satisfied: nvidia-nvtx-cu12==12.1.105 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (12.1.105)\n",
      "Requirement already satisfied: triton==2.2.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from torch->flash-attn) (2.2.0)\n",
      "Requirement already satisfied: nvidia-nvjitlink-cu12 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from nvidia-cusolver-cu12==11.4.5.107->torch->flash-attn) (12.5.82)\n",
      "Requirement already satisfied: MarkupSafe>=2.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from jinja2->torch->flash-attn) (2.1.5)\n",
      "Requirement already satisfied: mpmath<1.4.0,>=1.1.0 in /system/conda/miniconda3/envs/cloudspace/lib/python3.10/site-packages (from sympy->torch->flash-attn) (1.3.0)\n",
      "\n",
      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.1.1\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.2\u001b[0m\n",
      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "%pip install bitsandbytes accelerate\n",
    "%pip install flash-attn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "# os.environ['HF_TOKEN'] = \"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\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[INFO] Using attention implementation: sdpa\n",
      "[INFO] Using model_id: google/gemma-2b-it\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "`config.hidden_act` is ignored, you should use `config.hidden_activation` instead.\n",
      "Gemma's activation function will be set to `gelu_pytorch_tanh`. Please, use\n",
      "`config.hidden_activation` if you want to override this behaviour.\n",
      "See https://github.com/huggingface/transformers/pull/29402 for more details.\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "fbf6d36bf5934b81858ab7aa112994ab",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import torch\n",
    "from transformers import AutoTokenizer, AutoModelForCausalLM\n",
    "from transformers.utils import is_flash_attn_2_available \n",
    "\n",
    "# 1. Create quantization config for smaller model loading (optional)\n",
    "# Requires !pip install bitsandbytes accelerate, see: https://github.com/TimDettmers/bitsandbytes, https://huggingface.co/docs/accelerate/\n",
    "# For models that require 4-bit quantization (use this if you have low GPU memory available)\n",
    "from transformers import BitsAndBytesConfig\n",
    "quantization_config = BitsAndBytesConfig(load_in_4bit=True,\n",
    "                                         bnb_4bit_compute_dtype=torch.float16)\n",
    "\n",
    "# Bonus: Setup Flash Attention 2 for faster inference, default to \"sdpa\" or \"scaled dot product attention\" if it's not available\n",
    "# Flash Attention 2 requires NVIDIA GPU compute capability of 8.0 or above, see: https://developer.nvidia.com/cuda-gpus\n",
    "# Requires !pip install flash-attn, see: https://github.com/Dao-AILab/flash-attention \n",
    "if (is_flash_attn_2_available()) and (torch.cuda.get_device_capability(0)[0] >= 8):\n",
    "  attn_implementation = \"flash_attention_2\"\n",
    "else:\n",
    "  attn_implementation = \"sdpa\"\n",
    "print(f\"[INFO] Using attention implementation: {attn_implementation}\")\n",
    "\n",
    "# 2. Pick a model we'd like to use (this will depend on how much GPU memory you have available)\n",
    "#model_id = \"google/gemma-7b-it\"\n",
    "model_id = model_id # (we already set this above)\n",
    "print(f\"[INFO] Using model_id: {model_id}\")\n",
    "\n",
    "# 3. Instantiate tokenizer (tokenizer turns text into numbers ready for the model) \n",
    "tokenizer = AutoTokenizer.from_pretrained(pretrained_model_name_or_path=model_id)\n",
    "\n",
    "# 4. Instantiate the model\n",
    "llm_model = AutoModelForCausalLM.from_pretrained(pretrained_model_name_or_path=model_id, \n",
    "                                                 torch_dtype=torch.float16, # datatype to use, we want float16\n",
    "                                                 quantization_config=quantization_config if use_quantization_config else None,\n",
    "                                                 low_cpu_mem_usage=False, # use full memory \n",
    "                                                 attn_implementation=attn_implementation) # which attention version to use\n",
    "\n",
    "if not use_quantization_config: # quantization takes care of device setting automatically, so if it's not used, send model to GPU \n",
    "    llm_model.to(\"cuda\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GemmaForCausalLM(\n",
       "  (model): GemmaModel(\n",
       "    (embed_tokens): Embedding(256000, 2048, padding_idx=0)\n",
       "    (layers): ModuleList(\n",
       "      (0-17): 18 x GemmaDecoderLayer(\n",
       "        (self_attn): GemmaSdpaAttention(\n",
       "          (q_proj): Linear(in_features=2048, out_features=2048, bias=False)\n",
       "          (k_proj): Linear(in_features=2048, out_features=256, bias=False)\n",
       "          (v_proj): Linear(in_features=2048, out_features=256, bias=False)\n",
       "          (o_proj): Linear(in_features=2048, out_features=2048, bias=False)\n",
       "          (rotary_emb): GemmaRotaryEmbedding()\n",
       "        )\n",
       "        (mlp): GemmaMLP(\n",
       "          (gate_proj): Linear(in_features=2048, out_features=16384, bias=False)\n",
       "          (up_proj): Linear(in_features=2048, out_features=16384, bias=False)\n",
       "          (down_proj): Linear(in_features=16384, out_features=2048, bias=False)\n",
       "          (act_fn): PytorchGELUTanh()\n",
       "        )\n",
       "        (input_layernorm): GemmaRMSNorm()\n",
       "        (post_attention_layernorm): GemmaRMSNorm()\n",
       "      )\n",
       "    )\n",
       "    (norm): GemmaRMSNorm()\n",
       "  )\n",
       "  (lm_head): Linear(in_features=2048, out_features=256000, bias=False)\n",
       ")"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "llm_model\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2506172416"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_model_num_params(model: torch.nn.Module):\n",
    "    return sum([param.numel() for param in model.parameters()])\n",
    "\n",
    "get_model_num_params(llm_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'model_mem_bytes': 5012354048, 'model_mem_mb': 4780.15, 'model_mem_gb': 4.67}"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_model_mem_size(model: torch.nn.Module):\n",
    "    \"\"\"\n",
    "    Get how much memory a PyTorch model takes up.\n",
    "\n",
    "    See: https://discuss.pytorch.org/t/gpu-memory-that-model-uses/56822\n",
    "    \"\"\"\n",
    "    # Get model parameters and buffer sizes\n",
    "    mem_params = sum([param.nelement() * param.element_size() for param in model.parameters()])\n",
    "    mem_buffers = sum([buf.nelement() * buf.element_size() for buf in model.buffers()])\n",
    "\n",
    "    # Calculate various model sizes\n",
    "    model_mem_bytes = mem_params + mem_buffers # in bytes\n",
    "    model_mem_mb = model_mem_bytes / (1024**2) # in megabytes\n",
    "    model_mem_gb = model_mem_bytes / (1024**3) # in gigabytes\n",
    "\n",
    "    return {\"model_mem_bytes\": model_mem_bytes,\n",
    "            \"model_mem_mb\": round(model_mem_mb, 2),\n",
    "            \"model_mem_gb\": round(model_mem_gb, 2)}\n",
    "\n",
    "get_model_mem_size(llm_model)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Input text:\n",
      "Create a 7-day beginner gym workout plan focused on weight loss. The plan should include exercises for each day of the week, and it should cover a mix of strength training and cardio. Please specify the exercises, sets, and reps for each workout. For male \n",
      "\n",
      "Prompt (formatted):\n",
      "<bos><start_of_turn>user\n",
      "Create a 7-day beginner gym workout plan focused on weight loss. The plan should include exercises for each day of the week, and it should cover a mix of strength training and cardio. Please specify the exercises, sets, and reps for each workout. For male<end_of_turn>\n",
      "<start_of_turn>model\n",
      "\n"
     ]
    }
   ],
   "source": [
    "input_text = \"Create a 7-day beginner gym workout plan focused on weight loss. The plan should include exercises for each day of the week, and it should cover a mix of strength training and cardio. Please specify the exercises, sets, and reps for each workout. For male \"\n",
    "print(f\"Input text:\\n{input_text}\")\n",
    "scores,indices=retrieve_relevant_resources(query=input_text,embeddings= embeddings,n_resources_to_return=10)\n",
    "retrieved_passages = \"\"\n",
    "for index in indices:\n",
    "    retrieved_passages += pages_and_chunks[index][\"text\"] + \" \"\n",
    "combined_input = f\"Query: {input_text}\\nRelevant Information: {retrieved_passages}\"\n",
    "# Create prompt template for instruction-tuned model\n",
    "dialogue_template = [\n",
    "    {\"role\": \"user\",\n",
    "     \"content\":input_text}\n",
    "]\n",
    "\n",
    "# Apply the chat template\n",
    "prompt = tokenizer.apply_chat_template(conversation=dialogue_template,\n",
    "                                       tokenize=False, # keep as raw text (not tokenized)\n",
    "                                       add_generation_prompt=True)\n",
    "print(f\"\\nPrompt (formatted):\\n{prompt}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "torch.cuda.empty_cache()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model input (tokenized):\n",
      "{'input_ids': tensor([[     2,      2,    106,   1645,    108,   4912,    476, 235248, 235324,\n",
      "         235290,   1311,  53070,  15185,  29639,   1780,  14779,    611,   5171,\n",
      "           4783, 235265,    714,   1780,   1412,   3707,  21426,    604,   1853,\n",
      "           1744,    576,    573,   2788, 235269,    578,    665,   1412,   3098,\n",
      "            476,   7345,    576,   7268,   4770,    578,  27921, 235265,   5651,\n",
      "          23730,    573,  21426, 235269,   9032, 235269,    578,  63926,    604,\n",
      "           1853,  29639, 235265,   1699,   9202,    107,    108,    106,   2516,\n",
      "            108]], device='cuda:0'), 'attention_mask': tensor([[1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
      "         1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1,\n",
      "         1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]], device='cuda:0')}\n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model output (tokens):\n",
      "tensor([     2,      2,    106,   1645,    108,   4912,    476, 235248, 235324,\n",
      "        235290,   1311,  53070,  15185,  29639,   1780,  14779,    611,   5171,\n",
      "          4783, 235265,    714,   1780,   1412,   3707,  21426,    604,   1853,\n",
      "          1744,    576,    573,   2788, 235269,    578,    665,   1412,   3098,\n",
      "           476,   7345,    576,   7268,   4770,    578,  27921, 235265,   5651,\n",
      "         23730,    573,  21426, 235269,   9032, 235269,    578,  63926,    604,\n",
      "          1853,  29639, 235265,   1699,   9202,    107,    108,    106,   2516,\n",
      "           108,    688,   7366, 235248, 235274, 235292,  47085,    578,   6271,\n",
      "        101857,    688,    109, 235287,  98118,  12703,  47085,   5471, 235292,\n",
      "        235248, 235304,   9032,    576, 235248, 235274, 235276, 235290, 235274,\n",
      "        235284,  63926,    108, 235287,  98118,  12703,  20830,    484, 235292,\n",
      "        235248, 235304,   9032,    576, 235248, 235274, 235276, 235290, 235274,\n",
      "        235284,  63926,    108, 235287,   6271, 101857,  33680,  35064, 235292,\n",
      "        235248, 235304,   9032,    576, 235248, 235274, 235276, 235290, 235274,\n",
      "        235284,  63926,    109,    688,   7366, 235248, 235284, 235292,   6980,\n",
      "           578,  63220,   9268,    688,    109, 235287,   3053,  12703, 131223,\n",
      "        235292, 235248, 235304,   9032,    576, 235248, 235274, 235276, 235290,\n",
      "        235274, 235284,  63926,    108, 235287,  98118,  12703,  16138, 114448,\n",
      "        235292, 235248, 235304,   9032,    576, 235248, 235274, 235276, 235290,\n",
      "        235274, 235284,  63926,    108, 235287,  46013,   8624,   1560, 235292,\n",
      "        235248, 235304,   9032,    576, 235248, 235274, 235276, 235290, 235274,\n",
      "        235284,  63926,    109,    688,   7366, 235248, 235304, 235292,  98754,\n",
      "           578,  11953,    688,    109, 235287,  40557,   1989, 235292, 235248,\n",
      "        235304,   9032,    576, 235248, 235274, 235276, 235290, 235274, 235284,\n",
      "         63926,    108, 235287,  16138, 114448, 235292, 235248, 235304,   9032,\n",
      "           576, 235248, 235274, 235276, 235290, 235274, 235284,  63926,    108,\n",
      "        235287,  54490,    484, 235292, 235248, 235304,   9032,    576, 235248,\n",
      "        235274, 235276, 235290, 235274, 235284,  63926,    108, 235287, 153668,\n",
      "        235292, 235248, 235304,   9032,    576, 235248, 235304, 235276, 235290,\n",
      "        235318, 235276,  10102,    109,    688,   7366, 235248, 235310, 235292,\n",
      "        231224,    578,  13361,    688,    109, 235287, 149945,   5471, 235292,\n",
      "        235248, 235304,   9032,    576, 235248, 235274, 235276, 235290, 235274,\n",
      "        235284,  63926,    108, 235287,  97034, 175579, 235292, 235248, 235304,\n",
      "          9032,    576, 235248, 235274, 235276, 235290, 235274, 235284,  63926,\n",
      "           108, 235287,  19453, 175579, 235292, 235248, 235304,   9032,    576,\n",
      "        235248, 235274, 235276, 235290, 235274], device='cuda:0')\n",
      "\n"
     ]
    }
   ],
   "source": [
    "\n",
    "# Tokenize the input text (turn it into numbers) and send it to GPU\n",
    "input_ids = tokenizer(prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "print(f\"Model input (tokenized):\\n{input_ids}\\n\")\n",
    "\n",
    "# Generate outputs passed on the tokenized input\n",
    "# See generate docs: https://huggingface.co/docs/transformers/v4.38.2/en/main_classes/text_generation#transformers.GenerationConfig \n",
    "outputs = llm_model.generate(**input_ids,\n",
    "                             max_new_tokens=256) # define the maximum number of new tokens to create\n",
    "print(f\"Model output (tokens):\\n{outputs[0]}\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model output (decoded):\n",
      "<bos><bos><start_of_turn>user\n",
      "Create a 7-day beginner gym workout plan focused on weight loss. The plan should include exercises for each day of the week, and it should cover a mix of strength training and cardio. Please specify the exercises, sets, and reps for each workout. For male<end_of_turn>\n",
      "<start_of_turn>model\n",
      "**Day 1: Chest and Triceps**\n",
      "\n",
      "* Dumbbell Chest Press: 3 sets of 10-12 reps\n",
      "* Dumbbell Flyes: 3 sets of 10-12 reps\n",
      "* Triceps Pushdowns: 3 sets of 10-12 reps\n",
      "\n",
      "**Day 2: Back and Biceps**\n",
      "\n",
      "* Barbell Rows: 3 sets of 10-12 reps\n",
      "* Dumbbell Deadlifts: 3 sets of 10-12 reps\n",
      "* Hammer Curls: 3 sets of 10-12 reps\n",
      "\n",
      "**Day 3: Legs and Abs**\n",
      "\n",
      "* Squats: 3 sets of 10-12 reps\n",
      "* Deadlifts: 3 sets of 10-12 reps\n",
      "* Lunges: 3 sets of 10-12 reps\n",
      "* Plank: 3 sets of 30-60 seconds\n",
      "\n",
      "**Day 4: Shoulders and Core**\n",
      "\n",
      "* Overhead Press: 3 sets of 10-12 reps\n",
      "* Lateral Raises: 3 sets of 10-12 reps\n",
      "* Front Raises: 3 sets of 10-1\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Decode the output tokens to text\n",
    "outputs_decoded = tokenizer.decode(outputs[0])\n",
    "print(f\"Model output (decoded):\\n{outputs_decoded}\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}