centroid_movement.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from bertopic import BERTopic\n",
    "from bertopic.representation import PartOfSpeech,MaximalMarginalRelevance,KeyBERTInspired\n",
    "from sentence_transformers import SentenceTransformer\n",
    "import joblib\n",
    "from typing import List, Dict\n",
    "from sentence_transformers import SentenceTransformer, util\n",
    "import numpy as np\n",
    "import torch\n",
    "import json\n",
    "import scipy\n",
    "from src.utils import clean_dataset,list_to_dict,convert_ctfidf,NpEncoder\n",
    "\n",
    "model = SentenceTransformer('all-MiniLM-L6-v2')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Hello this is an example text with an email address Please remove punctuation and escape sequences including\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 18846/18846 [00:01<00:00, 12032.51it/s]\n"
     ]
    }
   ],
   "source": [
    "from sklearn.datasets import fetch_20newsgroups\n",
    "newsgroups_train = fetch_20newsgroups(subset='all')\n",
    "\n",
    "import re\n",
    "\n",
    "def load_model(model:int) :\n",
    "    \"\"\"Load the corresponding model based on the model number\n",
    "\n",
    "    Args:\n",
    "        model (int): ranging from 1 - 4 \n",
    "    \"\"\"\n",
    "    if model == 1 :\n",
    "        return BERTopic()\n",
    "    elif model == 2 :\n",
    "        representation_model = KeyBERTInspired()\n",
    "        return BERTopic(representation_model=representation_model)\n",
    "    elif model == 3 :\n",
    "        representation_model = PartOfSpeech(\"en_core_web_sm\")\n",
    "        return BERTopic(representation_model=representation_model)\n",
    "    else : \n",
    "        representation_model = MaximalMarginalRelevance(diversity=0.3)\n",
    "        return BERTopic(representation_model=representation_model)\n",
    "\n",
    "def clean_text(text):\n",
    "    # Remove escape sequences\n",
    "    text = re.sub(r'\\\\[a-zA-Z]', ' ', text)\n",
    "\n",
    "    # Remove email addresses\n",
    "    text = re.sub(r'\\S+@\\S+', ' ', text)\n",
    "\n",
    "    # Remove punctuation\n",
    "    text = re.sub(r'[^\\w\\s]', ' ', text)\n",
    "\n",
    "    # Remove extra whitespace\n",
    "    text = ' '.join(text.split())\n",
    "\n",
    "    return text\n",
    "\n",
    "# Example usage:\n",
    "text = \"Hello, this is an example text with an email address john@example.com. \\nPlease remove punctuation and escape sequences, including \\\\n.\"\n",
    "cleaned_text = clean_text(text)\n",
    "print(cleaned_text)\n",
    "\n",
    "\n",
    "from tqdm import tqdm\n",
    "\n",
    "docs = []\n",
    "for text in tqdm(newsgroups_train[\"data\"]):\n",
    "  docs.append(clean_text(text))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 18846/18846 [00:01<00:00, 12046.65it/s]\n",
      "/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/pynndescent/pynndescent_.py:346: NumbaPendingDeprecationWarning: \u001b[1m\u001b[1mCode using Numba extension API maybe depending on 'old_style' error-capturing, which is deprecated and will be replaced by 'new_style' in a future release. See details at https://numba.readthedocs.io/en/latest/reference/deprecation.html#deprecation-of-old-style-numba-captured-errors\n",
      "Exception origin:\n",
      "  File \"/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/numba/core/types/functions.py\", line 486, in __getnewargs__\n",
      "    raise ReferenceError(\"underlying object has vanished\")\n",
      "\u001b[0m\u001b[0m\n",
      "  init_rp_tree(data, dist, current_graph, leaf_array)\n",
      "/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/pynndescent/pynndescent_.py:348: NumbaPendingDeprecationWarning: \u001b[1m\u001b[1mCode using Numba extension API maybe depending on 'old_style' error-capturing, which is deprecated and will be replaced by 'new_style' in a future release. See details at https://numba.readthedocs.io/en/latest/reference/deprecation.html#deprecation-of-old-style-numba-captured-errors\n",
      "Exception origin:\n",
      "  File \"/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/numba/core/types/functions.py\", line 486, in __getnewargs__\n",
      "    raise ReferenceError(\"underlying object has vanished\")\n",
      "\u001b[0m\u001b[0m\n",
      "  init_random(n_neighbors, data, current_graph, dist, rng_state)\n",
      "/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/pynndescent/pynndescent_.py:358: NumbaPendingDeprecationWarning: \u001b[1m\u001b[1m\u001b[1mCode using Numba extension API maybe depending on 'old_style' error-capturing, which is deprecated and will be replaced by 'new_style' in a future release. See details at https://numba.readthedocs.io/en/latest/reference/deprecation.html#deprecation-of-old-style-numba-captured-errors\n",
      "Exception origin:\n",
      "  File \"/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/numba/core/types/functions.py\", line 486, in __getnewargs__\n",
      "    raise ReferenceError(\"underlying object has vanished\")\n",
      "\u001b[0m\u001b[0m\u001b[0m\n",
      "  nn_descent_internal_low_memory_parallel(\n",
      "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",
      "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",
      "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",
      "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",
      "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"
     ]
    },
    {
     "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>Topic</th>\n",
       "      <th>Count</th>\n",
       "      <th>Name</th>\n",
       "      <th>Representation</th>\n",
       "      <th>Representative_Docs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-1</td>\n",
       "      <td>6412</td>\n",
       "      <td>-1_to_the_of_and</td>\n",
       "      <td>[to, the, of, and, for, is, in, from, you, it]</td>\n",
       "      <td>[From Matt Freivald x8043 Subject Re Good Neig...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>388</td>\n",
       "      <td>0_pub_graphics_mail_edu</td>\n",
       "      <td>[pub, graphics, mail, edu, ftp, 128, siggraph,...</td>\n",
       "      <td>[From Dan Wallach Subject FAQ Typing Injuries ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>262</td>\n",
       "      <td>1_windows_dos_memory_swap</td>\n",
       "      <td>[windows, dos, memory, swap, disk, file, emm38...</td>\n",
       "      <td>[From Andrew Tong Subject Re Challenge to Micr...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2</td>\n",
       "      <td>189</td>\n",
       "      <td>2_cramer_gay_clayton_sexual</td>\n",
       "      <td>[cramer, gay, clayton, sexual, homosexual, men...</td>\n",
       "      <td>[From Clayton Cramer Subject Re Median Re New ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3</td>\n",
       "      <td>158</td>\n",
       "      <td>3_image_3d_graphics_phigs</td>\n",
       "      <td>[image, 3d, graphics, phigs, package, library,...</td>\n",
       "      <td>[From David B Lewis Subject comp windows x Fre...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366</th>\n",
       "      <td>365</td>\n",
       "      <td>11</td>\n",
       "      <td>365_tga_rle_pov_tmp</td>\n",
       "      <td>[tga, rle, pov, tmp, povray, convert, xloadima...</td>\n",
       "      <td>[Subject Re POVray tga rle From Craig Andrew H...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>367</th>\n",
       "      <td>366</td>\n",
       "      <td>11</td>\n",
       "      <td>366_funny_tsk_rahe_newbie</td>\n",
       "      <td>[funny, tsk, rahe, newbie, ipser, pettefar, te...</td>\n",
       "      <td>[From Pete Young Subject Re For 3 Last Organiz...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>368</th>\n",
       "      <td>367</td>\n",
       "      <td>11</td>\n",
       "      <td>367_solar_sail_sails_flexible</td>\n",
       "      <td>[solar, sail, sails, flexible, project, sunlig...</td>\n",
       "      <td>[From Bill Higgins Beam Jockey Subject Re Sola...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>369</th>\n",
       "      <td>368</td>\n",
       "      <td>11</td>\n",
       "      <td>368_freeman_censorship_garfiel_company</td>\n",
       "      <td>[freeman, censorship, garfiel, company, fraud,...</td>\n",
       "      <td>[From D DANIEL SOKOL Subject Re Enough Freeman...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>370</th>\n",
       "      <td>369</td>\n",
       "      <td>10</td>\n",
       "      <td>369_finland_finnish_players_nhl</td>\n",
       "      <td>[finland, finnish, players, nhl, team, howl, g...</td>\n",
       "      <td>[From Marko Poutiainen Subject Re Finland Swed...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>371 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Topic  Count                                    Name  \\\n",
       "0       -1   6412                        -1_to_the_of_and   \n",
       "1        0    388                 0_pub_graphics_mail_edu   \n",
       "2        1    262               1_windows_dos_memory_swap   \n",
       "3        2    189             2_cramer_gay_clayton_sexual   \n",
       "4        3    158               3_image_3d_graphics_phigs   \n",
       "..     ...    ...                                     ...   \n",
       "366    365     11                     365_tga_rle_pov_tmp   \n",
       "367    366     11               366_funny_tsk_rahe_newbie   \n",
       "368    367     11           367_solar_sail_sails_flexible   \n",
       "369    368     11  368_freeman_censorship_garfiel_company   \n",
       "370    369     10         369_finland_finnish_players_nhl   \n",
       "\n",
       "                                        Representation  \\\n",
       "0       [to, the, of, and, for, is, in, from, you, it]   \n",
       "1    [pub, graphics, mail, edu, ftp, 128, siggraph,...   \n",
       "2    [windows, dos, memory, swap, disk, file, emm38...   \n",
       "3    [cramer, gay, clayton, sexual, homosexual, men...   \n",
       "4    [image, 3d, graphics, phigs, package, library,...   \n",
       "..                                                 ...   \n",
       "366  [tga, rle, pov, tmp, povray, convert, xloadima...   \n",
       "367  [funny, tsk, rahe, newbie, ipser, pettefar, te...   \n",
       "368  [solar, sail, sails, flexible, project, sunlig...   \n",
       "369  [freeman, censorship, garfiel, company, fraud,...   \n",
       "370  [finland, finnish, players, nhl, team, howl, g...   \n",
       "\n",
       "                                   Representative_Docs  \n",
       "0    [From Matt Freivald x8043 Subject Re Good Neig...  \n",
       "1    [From Dan Wallach Subject FAQ Typing Injuries ...  \n",
       "2    [From Andrew Tong Subject Re Challenge to Micr...  \n",
       "3    [From Clayton Cramer Subject Re Median Re New ...  \n",
       "4    [From David B Lewis Subject comp windows x Fre...  \n",
       "..                                                 ...  \n",
       "366  [Subject Re POVray tga rle From Craig Andrew H...  \n",
       "367  [From Pete Young Subject Re For 3 Last Organiz...  \n",
       "368  [From Bill Higgins Beam Jockey Subject Re Sola...  \n",
       "369  [From D DANIEL SOKOL Subject Re Enough Freeman...  \n",
       "370  [From Marko Poutiainen Subject Re Finland Swed...  \n",
       "\n",
       "[371 rows x 5 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "docs = []\n",
    "count = 0\n",
    "for text in tqdm(newsgroups_train[\"data\"]):\n",
    "  count+=1\n",
    "  docs.append(clean_text(text) + f\" c_{count}\")\n",
    "\n",
    "topic_model = BERTopic()\n",
    "topics, probs = topic_model.fit_transform(docs)\n",
    "topic_model.get_topic_info()\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def centroid_ablation(word:str,document:str) -> float:\n",
    "    \"\"\"Take word out from document and see how much document embedding moves. \n",
    "\n",
    "    Args:\n",
    "        word (str): the input word\n",
    "        document (str): the input document \n",
    "\n",
    "    Returns:\n",
    "        float: the movement of the document embedding. \n",
    "    \"\"\"\n",
    "    document_embed = model.encode(document, convert_to_tensor=True)\n",
    "    document_without_word = document.replace(word, \"\")\n",
    "    document_without_word_embed = model.encode(document_without_word, convert_to_tensor=True)\n",
    "    return float(torch.linalg.norm(document_embed-document_without_word_embed))\n",
    "    \n",
    "# centroid_ablation(\"The\",\"The bird is on the tree. The cat is eating the bird.\")\n",
    "\n",
    "\n",
    "def get_spearman_rho(ctf_idf_json_topic1, ctf_idf_json_topic2):\n",
    "    # print(ctf_idf_json_topic1, ctf_idf_json_topic2)\n",
    "    words1 = list(ctf_idf_json_topic1.keys())\n",
    "    ctf_idf_rankings1 = [ctf_idf_json_topic1[word] for word in words1]\n",
    "    # print(f'Numbers for key {i}: {numbers_for_key}')\n",
    "    x = ctf_idf_rankings1\n",
    "\n",
    "    ctf_idf_rankings2 = [ctf_idf_json_topic2[word] for word in words1]\n",
    "    # print(f'Numbers for key {i}: {numbers_for_key}')\n",
    "    y = ctf_idf_rankings2\n",
    "    rho, p_value = scipy.stats.spearmanr(x, y)\n",
    "    return rho, p_value\n",
    "\n",
    "def get_kendall_tau(ctf_idf_json_topic1, ctf_idf_json_topic2):\n",
    "    # print(ctf_idf_json_topic1, ctf_idf_json_topic2)\n",
    "    words1 = list(ctf_idf_json_topic1.keys())\n",
    "    ctf_idf_rankings1 = [ctf_idf_json_topic1[word] for word in words1]\n",
    "    # print(f'Numbers for key {i}: {numbers_for_key}')\n",
    "    x = ctf_idf_rankings1\n",
    "\n",
    "    ctf_idf_rankings2 = [ctf_idf_json_topic2[word] for word in words1]\n",
    "    # print(f'Numbers for key {i}: {numbers_for_key}')\n",
    "    y = ctf_idf_rankings2\n",
    "    rho, p_value = scipy.stats.kendalltau(x, y)\n",
    "    return rho, p_value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "def raw_centroid_checks( docs: List[str], k: int, model: int) -> pd.DataFrame:\n",
    "    \"\"\"\n",
    "    Take as input a list of documents, a speicific number of topics to run the check for,\n",
    "    and the model. \n",
    "    1. Run a basic bertopic model to fit_transform the dataset.\n",
    "    2. For each topic in 1 to k :\n",
    "        2.1 Remove each repn word. \n",
    "        2.2 Check how much on avg the document embeddings change. \n",
    "    \"\"\"\n",
    "    final_ablation_mappings = {}\n",
    "    anchor_topic_model = load_model(model)\n",
    "    topics, probs = anchor_topic_model.fit_transform(docs)\n",
    "    topic_list = anchor_topic_model.get_topic_info()\n",
    "    c_tf_idf_mappings = anchor_topic_model.topic_representations_\n",
    "    c_tf_idf_mappings = convert_ctfidf(c_tf_idf_mappings)\n",
    "\n",
    "    documents = anchor_topic_model.get_topic_info()[\"Representative_Docs\"]\n",
    "\n",
    "    df_basic_mapping = pd.DataFrame({\"Document\": docs, \"Topic\": topics})\n",
    "\n",
    "    for topic_i in tqdm(range(k)):\n",
    "        topic_documents = df_basic_mapping[df_basic_mapping['Topic'] == topic_i][\"Document\"]\n",
    "        ablation_mappings = {}\n",
    "        for word in c_tf_idf_mappings[topic_i].keys():\n",
    "            centroid_movement = []\n",
    "            for document in topic_documents :\n",
    "                centroid_movement.append(centroid_ablation(word,document))\n",
    "            ablation_mappings[word] = np.mean(centroid_movement)\n",
    "        final_ablation_mappings[topic_i] = get_kendall_tau(c_tf_idf_mappings[topic_i],ablation_mappings)\n",
    "    return final_ablation_mappings\n",
    "\n",
    "# raw_centroid_checks(docs,2,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def run_centroid_ablation( docs: List[str], k: int, model: int, repeats: int, save_path: str)->None:\n",
    "    \"\"\"_summary_\n",
    "\n",
    "    Args:\n",
    "        docs (List[str]): _description_\n",
    "        k (int): _description_\n",
    "        model (int): _description_\n",
    "        repeats (int): _description_\n",
    "        save_path (str): _description_\n",
    "    \"\"\"\n",
    "\n",
    "    to_save = {}\n",
    "    for _ in range(repeats) : \n",
    "        to_save[_] = raw_centroid_checks(docs,k,model)\n",
    "    \n",
    "    with open(save_path, 'w') as json_file:\n",
    "        json.dump(to_save, json_file)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [09:39<00:00,  5.79s/it]\n"
     ]
    }
   ],
   "source": [
    "run_centroid_ablation(docs = docs , k = 100, model = 1, repeats = 1, save_path= \"final_result_kendall/centroid/centroid_1.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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",
      "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",
      "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",
      "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",
      "100%|██████████| 100/100 [10:51<00:00,  6.52s/it]\n"
     ]
    }
   ],
   "source": [
    "run_centroid_ablation(docs = docs , k = 100, model = 2, repeats = 1, save_path= \"final_result_kendall/centroid/centroid_2.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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",
      "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",
      "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",
      "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",
      "100%|██████████| 100/100 [10:35<00:00,  6.36s/it]\n"
     ]
    }
   ],
   "source": [
    "run_centroid_ablation(docs = docs , k = 100, model = 3, repeats = 1, save_path= \"final_result_kendall/centroid/centroid_3.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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",
      "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",
      "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",
      "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",
      "100%|██████████| 100/100 [10:53<00:00,  6.53s/it]\n"
     ]
    }
   ],
   "source": [
    "run_centroid_ablation(docs = docs , k = 100, model = 4, repeats = 1, save_path= \"final_result_kendall/centroid/centroid_4.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def read_data(path):\n",
    "    df = pd.read_csv(path)\n",
    "    return df['text'].tolist()\n",
    "\n",
    "nyt = read_data(\"data/nyt2020.csv\")\n",
    "wiki = read_data(\"data/wiki_en_10000.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/pynndescent/pynndescent_.py:346: NumbaPendingDeprecationWarning: \u001b[1m\u001b[1mCode using Numba extension API maybe depending on 'old_style' error-capturing, which is deprecated and will be replaced by 'new_style' in a future release. See details at https://numba.readthedocs.io/en/latest/reference/deprecation.html#deprecation-of-old-style-numba-captured-errors\n",
      "Exception origin:\n",
      "  File \"/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/numba/core/types/functions.py\", line 486, in __getnewargs__\n",
      "    raise ReferenceError(\"underlying object has vanished\")\n",
      "\u001b[0m\u001b[0m\n",
      "  init_rp_tree(data, dist, current_graph, leaf_array)\n",
      "/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/pynndescent/pynndescent_.py:348: NumbaPendingDeprecationWarning: \u001b[1m\u001b[1mCode using Numba extension API maybe depending on 'old_style' error-capturing, which is deprecated and will be replaced by 'new_style' in a future release. See details at https://numba.readthedocs.io/en/latest/reference/deprecation.html#deprecation-of-old-style-numba-captured-errors\n",
      "Exception origin:\n",
      "  File \"/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/numba/core/types/functions.py\", line 486, in __getnewargs__\n",
      "    raise ReferenceError(\"underlying object has vanished\")\n",
      "\u001b[0m\u001b[0m\n",
      "  init_random(n_neighbors, data, current_graph, dist, rng_state)\n",
      "/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/pynndescent/pynndescent_.py:358: NumbaPendingDeprecationWarning: \u001b[1m\u001b[1m\u001b[1mCode using Numba extension API maybe depending on 'old_style' error-capturing, which is deprecated and will be replaced by 'new_style' in a future release. See details at https://numba.readthedocs.io/en/latest/reference/deprecation.html#deprecation-of-old-style-numba-captured-errors\n",
      "Exception origin:\n",
      "  File \"/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/numba/core/types/functions.py\", line 486, in __getnewargs__\n",
      "    raise ReferenceError(\"underlying object has vanished\")\n",
      "\u001b[0m\u001b[0m\u001b[0m\n",
      "  nn_descent_internal_low_memory_parallel(\n",
      "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",
      "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",
      "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",
      "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",
      "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",
      "100%|██████████| 100/100 [08:35<00:00,  5.16s/it]\n",
      "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",
      "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",
      "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",
      "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",
      " 28%|██▊       | 28/100 [05:09<08:17,  6.91s/it]/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/scipy/stats/_stats_py.py:5445: ConstantInputWarning: An input array is constant; the correlation coefficient is not defined.\n",
      "  warnings.warn(stats.ConstantInputWarning(warn_msg))\n",
      "100%|██████████| 100/100 [09:38<00:00,  5.78s/it]\n",
      "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",
      "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",
      "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",
      "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",
      "100%|██████████| 100/100 [09:34<00:00,  5.75s/it]\n",
      "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",
      "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",
      "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",
      "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",
      "100%|██████████| 100/100 [09:29<00:00,  5.70s/it]\n"
     ]
    }
   ],
   "source": [
    "run_centroid_ablation(docs = nyt , k = 100, model = 1, repeats = 1, save_path= \"final_result/centroid/nyt_centroid_1.json\")\n",
    "run_centroid_ablation(docs = nyt , k = 100, model = 2, repeats = 1, save_path= \"final_result/centroid/nyt_centroid_2.json\")\n",
    "run_centroid_ablation(docs = nyt , k = 100, model = 3, repeats = 1, save_path= \"final_result/centroid/nyt_centroid_3.json\")\n",
    "run_centroid_ablation(docs = nyt , k = 100, model = 4, repeats = 1, save_path= \"final_result/centroid/nyt_centroid_4.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      " 43%|████▎     | 43/100 [10:46<08:12,  8.64s/it] /home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/scipy/stats/_stats_py.py:5445: ConstantInputWarning: An input array is constant; the correlation coefficient is not defined.\n",
      "  warnings.warn(stats.ConstantInputWarning(warn_msg))\n",
      "100%|██████████| 100/100 [15:10<00:00,  9.10s/it]\n",
      "  1%|          | 1/100 [01:05<1:47:49, 65.35s/it]/home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/scipy/stats/_stats_py.py:5445: ConstantInputWarning: An input array is constant; the correlation coefficient is not defined.\n",
      "  warnings.warn(stats.ConstantInputWarning(warn_msg))\n",
      "100%|██████████| 100/100 [14:07<00:00,  8.48s/it]\n",
      "100%|██████████| 100/100 [14:44<00:00,  8.84s/it]\n",
      " 19%|█▉        | 19/100 [06:21<15:32, 11.51s/it] /home/abpal/miniconda3/envs/bertopic_env/lib/python3.11/site-packages/scipy/stats/_stats_py.py:5445: ConstantInputWarning: An input array is constant; the correlation coefficient is not defined.\n",
      "  warnings.warn(stats.ConstantInputWarning(warn_msg))\n",
      "100%|██████████| 100/100 [14:53<00:00,  8.94s/it]\n"
     ]
    }
   ],
   "source": [
    "run_centroid_ablation(docs = wiki , k = 100, model = 1, repeats = 1, save_path= \"final_result/centroid/wiki_centroid_1.json\")\n",
    "run_centroid_ablation(docs = wiki , k = 100, model = 2, repeats = 1, save_path= \"final_result/centroid/wiki_centroid_2.json\")\n",
    "run_centroid_ablation(docs = wiki , k = 100, model = 3, repeats = 1, save_path= \"final_result/centroid/wiki_centroid_3.json\")\n",
    "run_centroid_ablation(docs = wiki , k = 100, model = 4, repeats = 1, save_path= \"final_result/centroid/wiki_centroid_4.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import math\n",
    "\n",
    "def remove_nan_from_list(float_list):\n",
    "    return list(filter(lambda x: not math.isnan(x), float_list))\n",
    "\n",
    "def read_json_and_plot(filename):\n",
    "    # Read JSON file\n",
    "    with open(filename, 'r') as file:\n",
    "        data = json.load(file)\n",
    "    \n",
    "    # Extract x and y values\n",
    "    x_values = []\n",
    "    y_values = []\n",
    "    for key in data:\n",
    "        # Iterate through inner dictionary\n",
    "        inner_dict = data[key]\n",
    "        for inner_key in inner_dict:\n",
    "            x_values.append(inner_dict[inner_key][0])  # First number in each item\n",
    "            y_values.append(inner_dict[inner_key][1])  # Second number in each item\n",
    "\n",
    "    # Plot\n",
    "    # plt.figure(figsize=(8, 6))\n",
    "    # plt.plot(x_values, y_values, 'bo', label='Data Points')\n",
    "    # plt.title('Normal Plot')\n",
    "    # plt.xlabel('X values')\n",
    "    # plt.ylabel('Y values')\n",
    "    # plt.grid(True)\n",
    "    # plt.legend()\n",
    "\n",
    "    # Calculate mean and standard deviation\n",
    "    x_values = remove_nan_from_list(x_values)\n",
    "    y_values = remove_nan_from_list(y_values)\n",
    "    x_mean = np.mean(x_values)\n",
    "    y_mean = np.mean(y_values)\n",
    "    x_std = np.std(x_values)\n",
    "    y_std = np.std(y_values)\n",
    "    print(f\"Mean of X values: {x_mean}\")\n",
    "    print(f\"Mean of Y values: {y_mean}\")\n",
    "    print(f\"Standard deviation of X values: {x_std}\")\n",
    "    print(f\"Standard deviation of Y values: {y_std}\")\n",
    "\n",
    "    # plt.show()\n",
    "\n",
    "# Usage\n",
    "# read_json_and_plot('your_json_file.json')\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Mean of X values: 0.06900737276706027\n",
      "Mean of Y values: 0.39241681400110506\n",
      "Standard deviation of X values: 0.41942395959733564\n",
      "Standard deviation of Y values: 0.2952454713213345\n"
     ]
    }
   ],
   "source": [
    "read_json_and_plot(\"final_result/centroid/wiki_centroid_1.json\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x1000 with 4 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import json\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "from scipy.stats import norm\n",
    "\n",
    "def plot_gaussian_distributions(filenames):\n",
    "    fig, axs = plt.subplots(2, 2, figsize=(12, 10))\n",
    "\n",
    "    model_name_dict = {\n",
    "        0 : \"c-TF-IDF\",\n",
    "        1 : \"KeyBERT Inspired\",\n",
    "        2 : \"POS\",\n",
    "        3 : \"MMR\"\n",
    "    }\n",
    "    for i, filename in enumerate(filenames):\n",
    "        # Read JSON file\n",
    "        with open(filename, 'r') as file:\n",
    "            data = json.load(file)\n",
    "        \n",
    "        # Extract x values\n",
    "        x_values = []\n",
    "        for key in data:\n",
    "            inner_dict = data[key]\n",
    "            for inner_key in inner_dict:\n",
    "                x_values.append(inner_dict[inner_key][0])  # First number in each item\n",
    "\n",
    "        # Plot Gaussian distribution on X values\n",
    "        row = i // 2\n",
    "        col = i % 2\n",
    "        ax = axs[row, col]\n",
    "\n",
    "        ax.hist(x_values, bins=30, density=True, alpha=0.6, color='g')\n",
    "\n",
    "        # Fit a normal distribution to the data\n",
    "        mu, std = np.mean(x_values), np.std(x_values)\n",
    "        xmin, xmax = ax.get_xlim()\n",
    "        x = np.linspace(xmin, xmax, 100)\n",
    "        p = norm.pdf(x, mu, std)\n",
    "        ax.plot(x, p, 'k', linewidth=2)\n",
    "        title = f\"{model_name_dict[i]} Results : mu = {mu:.2f}, std = {std:.2f}\"\n",
    "        ax.set_title(title)\n",
    "\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "\n",
    "# Usage\n",
    "filenames = ['final_result_kendall/centroid/centroid_1.json', 'final_result_kendall/centroid/centroid_2.json','final_result_kendall/centroid/centroid_3.json','final_result_kendall/centroid/centroid_4.json',]\n",
    "plot_gaussian_distributions(filenames)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "bertopic_env",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}