Notebooks/Disorder_Detection_Model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "0eb51d70-a544-4626-84e6-ee0bfcbf4a86",
   "metadata": {},
   "source": [
    "## Veri Seti Oluştur (Sadece veri setlerinde değişiklik yapılırsa çalıştır!)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "9dd05219-a21a-406a-9d04-d5e7e4761372",
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import os\n",
    "import glob\n",
    "from tqdm import tqdm\n",
    "from sklearn.model_selection import train_test_split"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "dc90f9e8-bc02-453a-9b00-a3db481f7ceb",
   "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>Text</th>\n",
       "      <th>Label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Sıkışmış hissetmek (yerine tekrar giriş yok, y...</td>\n",
       "      <td>Disorder</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Yakın zamanda başka bir şehre taşındım ve nele...</td>\n",
       "      <td>Disorder</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Panik atak geçirmenin eşiğindeydim, sadece bun...</td>\n",
       "      <td>Disorder</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Geçen hafta maruz kalma terapimi yaparken soka...</td>\n",
       "      <td>Disorder</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Bazen özgüven ve benlik imajıyla çok mücadele ...</td>\n",
       "      <td>Disorder</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                Text     Label\n",
       "0  Sıkışmış hissetmek (yerine tekrar giriş yok, y...  Disorder\n",
       "1  Yakın zamanda başka bir şehre taşındım ve nele...  Disorder\n",
       "2  Panik atak geçirmenin eşiğindeydim, sadece bun...  Disorder\n",
       "3  Geçen hafta maruz kalma terapimi yaparken soka...  Disorder\n",
       "4  Bazen özgüven ve benlik imajıyla çok mücadele ...  Disorder"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "disorder_texts = []\n",
    "for expurgated_data_path in glob.glob('../Data/Expurgated_Data/Reddit/*'):\n",
    "    #file_name = expurgated_data_path.split('/')[-1].split('.')[0]\n",
    "    disorder_df = pd.read_excel(expurgated_data_path)\n",
    "    disorder_texts.extend(disorder_df['Translated_Text'].values)\n",
    "\n",
    "disorder_df = pd.DataFrame({\n",
    "    \"Text\": disorder_texts,\n",
    "    \"Label\": \"Disorder\"\n",
    "})\n",
    "\n",
    "disorder_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "96115bfc",
   "metadata": {},
   "source": [
    "# Normal Veri Kümesini Yükleme ve Birleştirme\n",
    "\n",
    "Bu kod, normal veri kümesini yükler, bir pickle dosyasından ek verileri okur ve her iki veri kümesini birleştirir. Aşağıda her bir adımın açıklamaları yer almaktadır:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "dab3c27d-49c4-4e68-80ac-7f56a731f787",
   "metadata": {},
   "outputs": [],
   "source": [
    "normal_data = pd.read_csv('../Data/Disorder_Detection_Datasets/normal_dataset_translated_chunk_1.csv')\n",
    "\n",
    "import pickle\n",
    "\n",
    "with open('./normal_text_disorder_detection.pkl', 'rb') as file:\n",
    "    normal_texts_generated = pickle.load(file)\n",
    "\n",
    "\n",
    "normal_texts_generated_df = pd.DataFrame({\n",
    "    \"Text\": \"LLM Augmented veri olduğu için direkt türkçe üzerinden üretilmiştir.\",\n",
    "    \"Label\": \"Normal\",\n",
    "    \"Translated_Text\": normal_texts_generated\n",
    "})\n",
    "\n",
    "\n",
    "normal_total_data = pd.concat([normal_data, normal_texts_generated_df], axis=0).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "84244c73-cdd2-4d08-b430-4d1b30f017e1",
   "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>Text</th>\n",
       "      <th>Label</th>\n",
       "      <th>Translated_Text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>So let's be clear here. I'm totally fine.\\n\\nI...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Yani burada açık olalım. Ben tamamen iyiyim. B...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>I’m a 35 year old man, the sole income of a ho...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Ben 35 yaşında bir adamım, 3 kişilik bir evin ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>My life is over I’m about to be 21 and I alrea...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Hayatım bitti 21 olmak üzereyim ve zaten ölü h...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>I just want to outlet. I’m running out of opti...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Sadece çıkış yapmak istiyorum. Seçeneklerim tü...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Every time I start to feel depressed (It comes...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Ne zaman depresif hissetmeye başlasam (Dalgala...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22729</th>\n",
       "      <td>LLM Augmented veri olduğu için direkt türkçe ü...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Yemek yapmak benim için bir terapi gibi, keyif...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22730</th>\n",
       "      <td>LLM Augmented veri olduğu için direkt türkçe ü...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Doğada vakit geçirmek beni her zaman rahatlatı...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22731</th>\n",
       "      <td>LLM Augmented veri olduğu için direkt türkçe ü...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Arkadaşlarımla vakit geçirmek beni mutlu ediyor.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22732</th>\n",
       "      <td>LLM Augmented veri olduğu için direkt türkçe ü...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>Hayat kısa, her anın tadını çıkarmak gerekiyor.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22733</th>\n",
       "      <td>LLM Augmented veri olduğu için direkt türkçe ü...</td>\n",
       "      <td>Normal</td>\n",
       "      <td>İyimser olmak her zaman işe yarıyor.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>22734 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    Text   Label  \\\n",
       "0      So let's be clear here. I'm totally fine.\\n\\nI...  Normal   \n",
       "1      I’m a 35 year old man, the sole income of a ho...  Normal   \n",
       "2      My life is over I’m about to be 21 and I alrea...  Normal   \n",
       "3      I just want to outlet. I’m running out of opti...  Normal   \n",
       "4      Every time I start to feel depressed (It comes...  Normal   \n",
       "...                                                  ...     ...   \n",
       "22729  LLM Augmented veri olduğu için direkt türkçe ü...  Normal   \n",
       "22730  LLM Augmented veri olduğu için direkt türkçe ü...  Normal   \n",
       "22731  LLM Augmented veri olduğu için direkt türkçe ü...  Normal   \n",
       "22732  LLM Augmented veri olduğu için direkt türkçe ü...  Normal   \n",
       "22733  LLM Augmented veri olduğu için direkt türkçe ü...  Normal   \n",
       "\n",
       "                                         Translated_Text  \n",
       "0      Yani burada açık olalım. Ben tamamen iyiyim. B...  \n",
       "1      Ben 35 yaşında bir adamım, 3 kişilik bir evin ...  \n",
       "2      Hayatım bitti 21 olmak üzereyim ve zaten ölü h...  \n",
       "3      Sadece çıkış yapmak istiyorum. Seçeneklerim tü...  \n",
       "4      Ne zaman depresif hissetmeye başlasam (Dalgala...  \n",
       "...                                                  ...  \n",
       "22729  Yemek yapmak benim için bir terapi gibi, keyif...  \n",
       "22730  Doğada vakit geçirmek beni her zaman rahatlatı...  \n",
       "22731   Arkadaşlarımla vakit geçirmek beni mutlu ediyor.  \n",
       "22732    Hayat kısa, her anın tadını çıkarmak gerekiyor.  \n",
       "22733               İyimser olmak her zaman işe yarıyor.  \n",
       "\n",
       "[22734 rows x 3 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "normal_total_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2c626474-bb69-441f-ae79-a9bfb6948a8a",
   "metadata": {},
   "outputs": [],
   "source": [
    "total_df = pd.DataFrame()\n",
    "\n",
    "total_df['Text'] = normal_total_data['Translated_Text']\n",
    "total_df['Label'] = normal_total_data['Label']\n",
    "\n",
    "total_df = pd.concat([total_df, disorder_df], axis=0).reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "c84784cf-e8bb-429f-a10c-572228ef0da3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Label\n",
       "Normal      22734\n",
       "Disorder     8850\n",
       "Name: count, dtype: int64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "total_df['Label'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "37b80dc3-37de-4447-a173-3e25aa56e6fc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Saved\n"
     ]
    }
   ],
   "source": [
    "total_df.dropna(inplace=True)\n",
    "\n",
    "total_df.to_excel('../Data/Disorder_Detection_Datasets/main_normal_text_generated_dataset.xlsx', index=False)\n",
    "print(\"Saved\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92b99c5a-b189-4f4a-9660-07d319517dd6",
   "metadata": {},
   "source": [
    "### Veri Setini HuggingFace'e Push et"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "6c673a08-a9c8-4497-a98d-efd8bbf8494b",
   "metadata": {},
   "outputs": [],
   "source": [
    "from datasets import Dataset, DatasetDict\n",
    "\n",
    "train, test = train_test_split(total_df, test_size = 0.3, random_state=42)\n",
    "\n",
    "train_data = Dataset.from_pandas(train, preserve_index=False)\n",
    "test_data = Dataset.from_pandas(test, preserve_index=False)\n",
    "\n",
    "hg_data = DatasetDict({\n",
    "    \"train\": train_data,\n",
    "    \"test\": test_data\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4691c365-72f9-41b1-8336-34b298b086c2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DatasetDict({\n",
       "    train: Dataset({\n",
       "        features: ['Text', 'Label'],\n",
       "        num_rows: 22108\n",
       "    })\n",
       "    test: Dataset({\n",
       "        features: ['Text', 'Label'],\n",
       "        num_rows: 9475\n",
       "    })\n",
       "})"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hg_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "dc2917ad-4260-4c27-a9d3-cceb691465e8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Token will not been saved to git credential helper. Pass `add_to_git_credential=True` if you want to set the git credential as well.\n",
      "Token is valid (permission: write).\n",
      "Your token has been saved to C:\\Users\\halilibrahim.hatun\\.cache\\huggingface\\token\n",
      "Login successful\n"
     ]
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "54309d8b54f64bc49fffc39196331e7c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Uploading the dataset shards:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "86ef0d8194894990a073bff56bbbe257",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Creating parquet from Arrow format:   0%|          | 0/23 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "365d888139a9483a84c1ea6a300f2310",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Uploading the dataset shards:   0%|          | 0/1 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "3ea33fff88aa4ef0afcc6d7cbcc028e2",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Creating parquet from Arrow format:   0%|          | 0/10 [00:00<?, ?ba/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "4724d30e273648648a89f2b6f2a0dd36",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "README.md:   0%|          | 0.00/408 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Data was pushed :)\n"
     ]
    }
   ],
   "source": [
    "!huggingface-cli login --token=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\n",
    "\n",
    "hg_data.push_to_hub(\"halilibr/dilbazlar-binary-disorder-detection-normal-generated-dataset\")\n",
    "print(\"Data was pushed :)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "8368ee72-0839-4e47-8da0-8b0627b0ee1a",
   "metadata": {},
   "source": [
    "## Eğitime Başla"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "be19ce61-5fb4-4333-8782-2ac6688e3a14",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Device:  cuda\n",
      "Token will not been saved to git credential helper. Pass `add_to_git_credential=True` if you want to set the git credential as well.\n",
      "Token is valid (permission: write).\n",
      "Your token has been saved to C:\\Users\\halilibrahim.hatun\\.cache\\huggingface\\token\n",
      "Login successful\n"
     ]
    }
   ],
   "source": [
    "import json\n",
    "import os\n",
    "from sklearn.metrics import roc_curve\n",
    "from sklearn.metrics import accuracy_score\n",
    "from sklearn.metrics import f1_score\n",
    "from sklearn.model_selection import train_test_split\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import classification_report\n",
    "import transformers\n",
    "from transformers import BertModel, BertTokenizer\n",
    "import warnings\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import glob\n",
    "from tqdm import tqdm\n",
    "\n",
    "# Huggingface\n",
    "from datasets import load_dataset\n",
    "import huggingface_hub\n",
    "\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "# specify GPU\n",
    "device = torch.device('cuda' if torch.cuda.is_available() else 'cpu')\n",
    "\n",
    "print(\"Device: \", device)\n",
    "\n",
    "!huggingface-cli login --token=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"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "caf81828-2333-4bf0-8c34-20b5528aa988",
   "metadata": {},
   "source": [
    "## Veri Setini Yükle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "bc0e4798-4378-442f-bac9-b50736b4b3e0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "972ae2ffd7ee4e72b1bb95653d399d87",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading readme:   0%|          | 0.00/408 [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a0134bf0bd024e7d869981a784332aee",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data:   0%|          | 0.00/2.09M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "051d5a98b75a41d3bf2cd7201d96490b",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Downloading data:   0%|          | 0.00/903k [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "0b3574f07d6042028f8e389b7c888e0c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Generating train split:   0%|          | 0/22108 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "ename": "DatasetGenerationError",
     "evalue": "An error occurred while generating the dataset",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mFileNotFoundError\u001b[0m                         Traceback (most recent call last)",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\datasets\\builder.py:1962\u001b[0m, in \u001b[0;36mArrowBasedBuilder._prepare_split_single\u001b[1;34m(self, gen_kwargs, fpath, file_format, max_shard_size, job_id)\u001b[0m\n\u001b[0;32m   1961\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m-> 1962\u001b[0m     writer \u001b[38;5;241m=\u001b[39m writer_class(\n\u001b[0;32m   1963\u001b[0m         features\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39minfo\u001b[38;5;241m.\u001b[39mfeatures,\n\u001b[0;32m   1964\u001b[0m         path\u001b[38;5;241m=\u001b[39mfpath\u001b[38;5;241m.\u001b[39mreplace(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mSSSSS\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mshard_id\u001b[38;5;132;01m:\u001b[39;00m\u001b[38;5;124m05d\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m)\u001b[38;5;241m.\u001b[39mreplace(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mJJJJJ\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mjob_id\u001b[38;5;132;01m:\u001b[39;00m\u001b[38;5;124m05d\u001b[39m\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m),\n\u001b[0;32m   1965\u001b[0m         writer_batch_size\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_writer_batch_size,\n\u001b[0;32m   1966\u001b[0m         storage_options\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fs\u001b[38;5;241m.\u001b[39mstorage_options,\n\u001b[0;32m   1967\u001b[0m         embed_local_files\u001b[38;5;241m=\u001b[39membed_local_files,\n\u001b[0;32m   1968\u001b[0m     )\n\u001b[0;32m   1969\u001b[0m     \u001b[38;5;28;01mtry\u001b[39;00m:\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\datasets\\arrow_writer.py:336\u001b[0m, in \u001b[0;36mArrowWriter.__init__\u001b[1;34m(self, schema, features, path, stream, fingerprint, writer_batch_size, hash_salt, check_duplicates, disable_nullable, update_features, with_metadata, unit, embed_local_files, storage_options)\u001b[0m\n\u001b[0;32m    331\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_path \u001b[38;5;241m=\u001b[39m (\n\u001b[0;32m    332\u001b[0m     fs_token_paths[\u001b[38;5;241m2\u001b[39m][\u001b[38;5;241m0\u001b[39m]\n\u001b[0;32m    333\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m is_remote_filesystem(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fs)\n\u001b[0;32m    334\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fs\u001b[38;5;241m.\u001b[39munstrip_protocol(fs_token_paths[\u001b[38;5;241m2\u001b[39m][\u001b[38;5;241m0\u001b[39m])\n\u001b[0;32m    335\u001b[0m )\n\u001b[1;32m--> 336\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstream \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_fs\u001b[38;5;241m.\u001b[39mopen(fs_token_paths[\u001b[38;5;241m2\u001b[39m][\u001b[38;5;241m0\u001b[39m], \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mwb\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m    337\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_closable_stream \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\fsspec\\spec.py:1307\u001b[0m, in \u001b[0;36mAbstractFileSystem.open\u001b[1;34m(self, path, mode, block_size, cache_options, compression, **kwargs)\u001b[0m\n\u001b[0;32m   1306\u001b[0m ac \u001b[38;5;241m=\u001b[39m kwargs\u001b[38;5;241m.\u001b[39mpop(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mautocommit\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_intrans)\n\u001b[1;32m-> 1307\u001b[0m f \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_open(\n\u001b[0;32m   1308\u001b[0m     path,\n\u001b[0;32m   1309\u001b[0m     mode\u001b[38;5;241m=\u001b[39mmode,\n\u001b[0;32m   1310\u001b[0m     block_size\u001b[38;5;241m=\u001b[39mblock_size,\n\u001b[0;32m   1311\u001b[0m     autocommit\u001b[38;5;241m=\u001b[39mac,\n\u001b[0;32m   1312\u001b[0m     cache_options\u001b[38;5;241m=\u001b[39mcache_options,\n\u001b[0;32m   1313\u001b[0m     \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs,\n\u001b[0;32m   1314\u001b[0m )\n\u001b[0;32m   1315\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m compression \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\fsspec\\implementations\\local.py:180\u001b[0m, in \u001b[0;36mLocalFileSystem._open\u001b[1;34m(self, path, mode, block_size, **kwargs)\u001b[0m\n\u001b[0;32m    179\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmakedirs(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_parent(path), exist_ok\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m)\n\u001b[1;32m--> 180\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m LocalFileOpener(path, mode, fs\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\fsspec\\implementations\\local.py:302\u001b[0m, in \u001b[0;36mLocalFileOpener.__init__\u001b[1;34m(self, path, mode, autocommit, fs, compression, **kwargs)\u001b[0m\n\u001b[0;32m    301\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mblocksize \u001b[38;5;241m=\u001b[39m io\u001b[38;5;241m.\u001b[39mDEFAULT_BUFFER_SIZE\n\u001b[1;32m--> 302\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_open()\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\fsspec\\implementations\\local.py:307\u001b[0m, in \u001b[0;36mLocalFileOpener._open\u001b[1;34m(self)\u001b[0m\n\u001b[0;32m    306\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mautocommit \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mw\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmode:\n\u001b[1;32m--> 307\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mf \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mopen\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mpath, mode\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmode)\n\u001b[0;32m    308\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcompression:\n",
      "\u001b[1;31mFileNotFoundError\u001b[0m: [Errno 2] No such file or directory: 'C:/Users/halilibrahim.hatun/.cache/huggingface/datasets/halilibr___dilbazlar-binary-disorder-detection-normal-generated-dataset/default/0.0.0/de5d806af50a42981b2606ad5df26901189144ba.incomplete/dilbazlar-binary-disorder-detection-normal-generated-dataset-train-00000-00000-of-NNNNN.arrow'",
      "\nThe above exception was the direct cause of the following exception:\n",
      "\u001b[1;31mDatasetGenerationError\u001b[0m                    Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[3], line 1\u001b[0m\n\u001b[1;32m----> 1\u001b[0m dataset \u001b[38;5;241m=\u001b[39m load_dataset(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mhalilibr/dilbazlar-binary-disorder-detection-normal-generated-dataset\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m      2\u001b[0m dataset[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtrain\u001b[39m\u001b[38;5;124m\"\u001b[39m][\u001b[38;5;241m100\u001b[39m]\n\u001b[0;32m      4\u001b[0m dataset \u001b[38;5;241m=\u001b[39m hg_data\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\datasets\\load.py:2545\u001b[0m, in \u001b[0;36mload_dataset\u001b[1;34m(path, name, data_dir, data_files, split, cache_dir, features, download_config, download_mode, verification_mode, ignore_verifications, keep_in_memory, save_infos, revision, token, use_auth_token, task, streaming, num_proc, storage_options, trust_remote_code, **config_kwargs)\u001b[0m\n\u001b[0;32m   2542\u001b[0m try_from_hf_gcs \u001b[38;5;241m=\u001b[39m path \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;129;01min\u001b[39;00m _PACKAGED_DATASETS_MODULES\n\u001b[0;32m   2544\u001b[0m \u001b[38;5;66;03m# Download and prepare data\u001b[39;00m\n\u001b[1;32m-> 2545\u001b[0m builder_instance\u001b[38;5;241m.\u001b[39mdownload_and_prepare(\n\u001b[0;32m   2546\u001b[0m     download_config\u001b[38;5;241m=\u001b[39mdownload_config,\n\u001b[0;32m   2547\u001b[0m     download_mode\u001b[38;5;241m=\u001b[39mdownload_mode,\n\u001b[0;32m   2548\u001b[0m     verification_mode\u001b[38;5;241m=\u001b[39mverification_mode,\n\u001b[0;32m   2549\u001b[0m     try_from_hf_gcs\u001b[38;5;241m=\u001b[39mtry_from_hf_gcs,\n\u001b[0;32m   2550\u001b[0m     num_proc\u001b[38;5;241m=\u001b[39mnum_proc,\n\u001b[0;32m   2551\u001b[0m     storage_options\u001b[38;5;241m=\u001b[39mstorage_options,\n\u001b[0;32m   2552\u001b[0m )\n\u001b[0;32m   2554\u001b[0m \u001b[38;5;66;03m# Build dataset for splits\u001b[39;00m\n\u001b[0;32m   2555\u001b[0m keep_in_memory \u001b[38;5;241m=\u001b[39m (\n\u001b[0;32m   2556\u001b[0m     keep_in_memory \u001b[38;5;28;01mif\u001b[39;00m keep_in_memory \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;28;01melse\u001b[39;00m is_small_dataset(builder_instance\u001b[38;5;241m.\u001b[39minfo\u001b[38;5;241m.\u001b[39mdataset_size)\n\u001b[0;32m   2557\u001b[0m )\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\datasets\\builder.py:1003\u001b[0m, in \u001b[0;36mDatasetBuilder.download_and_prepare\u001b[1;34m(self, output_dir, download_config, download_mode, verification_mode, ignore_verifications, try_from_hf_gcs, dl_manager, base_path, use_auth_token, file_format, max_shard_size, num_proc, storage_options, **download_and_prepare_kwargs)\u001b[0m\n\u001b[0;32m   1001\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m num_proc \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m   1002\u001b[0m         prepare_split_kwargs[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mnum_proc\u001b[39m\u001b[38;5;124m\"\u001b[39m] \u001b[38;5;241m=\u001b[39m num_proc\n\u001b[1;32m-> 1003\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_download_and_prepare(\n\u001b[0;32m   1004\u001b[0m         dl_manager\u001b[38;5;241m=\u001b[39mdl_manager,\n\u001b[0;32m   1005\u001b[0m         verification_mode\u001b[38;5;241m=\u001b[39mverification_mode,\n\u001b[0;32m   1006\u001b[0m         \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mprepare_split_kwargs,\n\u001b[0;32m   1007\u001b[0m         \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mdownload_and_prepare_kwargs,\n\u001b[0;32m   1008\u001b[0m     )\n\u001b[0;32m   1009\u001b[0m \u001b[38;5;66;03m# Sync info\u001b[39;00m\n\u001b[0;32m   1010\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39minfo\u001b[38;5;241m.\u001b[39mdataset_size \u001b[38;5;241m=\u001b[39m \u001b[38;5;28msum\u001b[39m(split\u001b[38;5;241m.\u001b[39mnum_bytes \u001b[38;5;28;01mfor\u001b[39;00m split \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39minfo\u001b[38;5;241m.\u001b[39msplits\u001b[38;5;241m.\u001b[39mvalues())\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\datasets\\builder.py:1098\u001b[0m, in \u001b[0;36mDatasetBuilder._download_and_prepare\u001b[1;34m(self, dl_manager, verification_mode, **prepare_split_kwargs)\u001b[0m\n\u001b[0;32m   1094\u001b[0m split_dict\u001b[38;5;241m.\u001b[39madd(split_generator\u001b[38;5;241m.\u001b[39msplit_info)\n\u001b[0;32m   1096\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m   1097\u001b[0m     \u001b[38;5;66;03m# Prepare split will record examples associated to the split\u001b[39;00m\n\u001b[1;32m-> 1098\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_prepare_split(split_generator, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mprepare_split_kwargs)\n\u001b[0;32m   1099\u001b[0m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mOSError\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[0;32m   1100\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mOSError\u001b[39;00m(\n\u001b[0;32m   1101\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mCannot find data file. \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m   1102\u001b[0m         \u001b[38;5;241m+\u001b[39m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mmanual_download_instructions \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m   1103\u001b[0m         \u001b[38;5;241m+\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124mOriginal error:\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m   1104\u001b[0m         \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mstr\u001b[39m(e)\n\u001b[0;32m   1105\u001b[0m     ) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\datasets\\builder.py:1858\u001b[0m, in \u001b[0;36mArrowBasedBuilder._prepare_split\u001b[1;34m(self, split_generator, file_format, num_proc, max_shard_size)\u001b[0m\n\u001b[0;32m   1856\u001b[0m job_id \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m\n\u001b[0;32m   1857\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m pbar:\n\u001b[1;32m-> 1858\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m job_id, done, content \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_prepare_split_single(\n\u001b[0;32m   1859\u001b[0m         gen_kwargs\u001b[38;5;241m=\u001b[39mgen_kwargs, job_id\u001b[38;5;241m=\u001b[39mjob_id, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39m_prepare_split_args\n\u001b[0;32m   1860\u001b[0m     ):\n\u001b[0;32m   1861\u001b[0m         \u001b[38;5;28;01mif\u001b[39;00m done:\n\u001b[0;32m   1862\u001b[0m             result \u001b[38;5;241m=\u001b[39m content\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\datasets\\builder.py:2014\u001b[0m, in \u001b[0;36mArrowBasedBuilder._prepare_split_single\u001b[1;34m(self, gen_kwargs, fpath, file_format, max_shard_size, job_id)\u001b[0m\n\u001b[0;32m   2012\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(e, DatasetGenerationError):\n\u001b[0;32m   2013\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m\n\u001b[1;32m-> 2014\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m DatasetGenerationError(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mAn error occurred while generating the dataset\u001b[39m\u001b[38;5;124m\"\u001b[39m) \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01me\u001b[39;00m\n\u001b[0;32m   2016\u001b[0m \u001b[38;5;28;01myield\u001b[39;00m job_id, \u001b[38;5;28;01mTrue\u001b[39;00m, (total_num_examples, total_num_bytes, writer\u001b[38;5;241m.\u001b[39m_features, num_shards, shard_lengths)\n",
      "\u001b[1;31mDatasetGenerationError\u001b[0m: An error occurred while generating the dataset"
     ]
    }
   ],
   "source": [
    "# dataset = load_dataset(\"halilibr/dilbazlar-binary-disorder-detection-normal-generated-dataset\")\n",
    "# dataset[\"train\"][100]\n",
    "\n",
    "dataset = hg_data "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "2fe0f254-a776-4c6b-90ee-0ead81f2c72b",
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = dataset.rename_column(\"Label\", \"labels\")\n",
    "dataset = dataset.rename_column(\"Text\", \"text\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "85e884dd-aca5-49e3-9c01-8f0f764baf39",
   "metadata": {},
   "source": [
    "## Tokenizer'i oluştur"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "b538893b-2b41-4a1c-b589-b387b59caa87",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "68f92f9657d04d5a8bb7c8e9e5b219a1",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Map:   0%|          | 0/22108 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "52d5cbde91f646c082b704e26df19939",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "Map:   0%|          | 0/9475 [00:00<?, ? examples/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from transformers import AutoTokenizer\n",
    "\n",
    "tokenizer = AutoTokenizer.from_pretrained(\"dbmdz/bert-base-turkish-128k-uncased\")\n",
    "\n",
    "\n",
    "def tokenize_function(examples):\n",
    "   label_mapping = {\n",
    "        \"Normal\": 0,\n",
    "        \"Disorder\": 1\n",
    "    }\n",
    "   inputs = tokenizer(examples[\"text\"], padding=\"max_length\", max_length=150, truncation=True)\n",
    "\n",
    "   labels = [label_mapping[label] for label in examples[\"labels\"]]\n",
    "   labels = torch.tensor(labels)\n",
    "\n",
    "   return {\n",
    "    \"input_ids\": inputs[\"input_ids\"],\n",
    "    \"attention_mask\": inputs[\"attention_mask\"],\n",
    "    \"labels\": labels,\n",
    "   }\n",
    "\n",
    "tokenized_dataset = dataset.map(tokenize_function, batched=True, batch_size=16)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "fa654b4c-f996-4dae-bb70-bd5c4c723082",
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import DataCollatorWithPadding\n",
    "\n",
    "data_collator = DataCollatorWithPadding(tokenizer=tokenizer)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a601c10c-ca72-47db-b249-f3c103a3e52c",
   "metadata": {},
   "source": [
    "## Trainer Yaklaşımı ile Eğitim"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "a6ff5552-9481-4a15-a796-ca952a82e897",
   "metadata": {},
   "outputs": [],
   "source": [
    "id2label = {\n",
    "    0: \"Normal\",\n",
    "    1: \"Disorder\"\n",
    "}\n",
    "label2id = {\n",
    "    \"Normal\": 0,\n",
    "    \"Disorder\": 1\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "474277ad-0375-4458-8998-50edb997a750",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of BertForTokenClassification were not initialized from the model checkpoint at dbmdz/bert-base-turkish-128k-uncased and are newly initialized: ['classifier.weight', 'classifier.bias']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    }
   ],
   "source": [
    "from transformers import AutoModelForTokenClassification \n",
    "\n",
    "model = AutoModelForTokenClassification.from_pretrained(\"dbmdz/bert-base-turkish-128k-uncased\", num_labels=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "2fb32c69-9eb9-4f93-b322-f6de256f17e7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import TrainingArguments\n",
    "\n",
    "training_args = TrainingArguments(output_dir=\"test_trainer\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6c09bf85-b898-4671-9cab-1927fdafce35",
   "metadata": {},
   "source": [
    "### Değerlendirme Metrikleri"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "43c6e59c-30bb-4e5f-a009-32d702e91020",
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import evaluate\n",
    "\n",
    "metric = evaluate.load(\"accuracy\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "c87d11fc-1315-440f-ba85-40ec64c3b1d6",
   "metadata": {},
   "outputs": [],
   "source": [
    "def compute_metrics(eval_pred):\n",
    "    logits, labels = eval_pred\n",
    "    predictions = np.argmax(logits, axis=-1)\n",
    "    return metric.compute(predictions=predictions, references=labels)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "f82af08e-cfa3-48ed-9f56-82b31031b12a",
   "metadata": {},
   "outputs": [],
   "source": [
    "from transformers import TrainingArguments, Trainer\n",
    "\n",
    "training_args = TrainingArguments(\n",
    "    output_dir=\"temp_disorder_detect\",\n",
    "    learning_rate=0.00005,\n",
    "    per_device_train_batch_size=16,\n",
    "    per_device_eval_batch_size=16,\n",
    "    num_train_epochs=5,\n",
    "    weight_decay=0.01,\n",
    "    evaluation_strategy=\"epoch\",\n",
    "    save_strategy=\"epoch\",\n",
    "    load_best_model_at_end=True,\n",
    "    save_total_limit=3,\n",
    "    report_to='tensorboard'\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "601baee0-3688-4e20-a03d-7c366af5bbe4",
   "metadata": {},
   "outputs": [],
   "source": [
    "trainer = Trainer(\n",
    "    model=model,\n",
    "    args=training_args,\n",
    "    train_dataset=train_tokenized_dataset,\n",
    "    eval_dataset=test_tokenized_dataset,\n",
    "    tokenizer=tokenizer,\n",
    "    data_collator=data_collator,\n",
    "    compute_metrics=compute_metrics,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "cb4fdbb2-94f5-4a59-8d69-444a4264b53e",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "You're using a BertTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n"
     ]
    },
    {
     "ename": "ValueError",
     "evalue": "Expected input batch_size (4400) to match target batch_size (16).",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[64], line 1\u001b[0m\n\u001b[1;32m----> 1\u001b[0m trainer\u001b[38;5;241m.\u001b[39mtrain()\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\transformers\\trainer.py:1537\u001b[0m, in \u001b[0;36mTrainer.train\u001b[1;34m(self, resume_from_checkpoint, trial, ignore_keys_for_eval, **kwargs)\u001b[0m\n\u001b[0;32m   1535\u001b[0m         hf_hub_utils\u001b[38;5;241m.\u001b[39menable_progress_bars()\n\u001b[0;32m   1536\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m-> 1537\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m inner_training_loop(\n\u001b[0;32m   1538\u001b[0m         args\u001b[38;5;241m=\u001b[39margs,\n\u001b[0;32m   1539\u001b[0m         resume_from_checkpoint\u001b[38;5;241m=\u001b[39mresume_from_checkpoint,\n\u001b[0;32m   1540\u001b[0m         trial\u001b[38;5;241m=\u001b[39mtrial,\n\u001b[0;32m   1541\u001b[0m         ignore_keys_for_eval\u001b[38;5;241m=\u001b[39mignore_keys_for_eval,\n\u001b[0;32m   1542\u001b[0m     )\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\transformers\\trainer.py:1854\u001b[0m, in \u001b[0;36mTrainer._inner_training_loop\u001b[1;34m(self, batch_size, args, resume_from_checkpoint, trial, ignore_keys_for_eval)\u001b[0m\n\u001b[0;32m   1851\u001b[0m     \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcontrol \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcallback_handler\u001b[38;5;241m.\u001b[39mon_step_begin(args, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstate, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcontrol)\n\u001b[0;32m   1853\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39maccelerator\u001b[38;5;241m.\u001b[39maccumulate(model):\n\u001b[1;32m-> 1854\u001b[0m     tr_loss_step \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mtraining_step(model, inputs)\n\u001b[0;32m   1856\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m (\n\u001b[0;32m   1857\u001b[0m     args\u001b[38;5;241m.\u001b[39mlogging_nan_inf_filter\n\u001b[0;32m   1858\u001b[0m     \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m is_torch_tpu_available()\n\u001b[0;32m   1859\u001b[0m     \u001b[38;5;129;01mand\u001b[39;00m (torch\u001b[38;5;241m.\u001b[39misnan(tr_loss_step) \u001b[38;5;129;01mor\u001b[39;00m torch\u001b[38;5;241m.\u001b[39misinf(tr_loss_step))\n\u001b[0;32m   1860\u001b[0m ):\n\u001b[0;32m   1861\u001b[0m     \u001b[38;5;66;03m# if loss is nan or inf simply add the average of previous logged losses\u001b[39;00m\n\u001b[0;32m   1862\u001b[0m     tr_loss \u001b[38;5;241m+\u001b[39m\u001b[38;5;241m=\u001b[39m tr_loss \u001b[38;5;241m/\u001b[39m (\u001b[38;5;241m1\u001b[39m \u001b[38;5;241m+\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mstate\u001b[38;5;241m.\u001b[39mglobal_step \u001b[38;5;241m-\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_globalstep_last_logged)\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\transformers\\trainer.py:2735\u001b[0m, in \u001b[0;36mTrainer.training_step\u001b[1;34m(self, model, inputs)\u001b[0m\n\u001b[0;32m   2732\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m loss_mb\u001b[38;5;241m.\u001b[39mreduce_mean()\u001b[38;5;241m.\u001b[39mdetach()\u001b[38;5;241m.\u001b[39mto(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39margs\u001b[38;5;241m.\u001b[39mdevice)\n\u001b[0;32m   2734\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcompute_loss_context_manager():\n\u001b[1;32m-> 2735\u001b[0m     loss \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mcompute_loss(model, inputs)\n\u001b[0;32m   2737\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39margs\u001b[38;5;241m.\u001b[39mn_gpu \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m1\u001b[39m:\n\u001b[0;32m   2738\u001b[0m     loss \u001b[38;5;241m=\u001b[39m loss\u001b[38;5;241m.\u001b[39mmean()  \u001b[38;5;66;03m# mean() to average on multi-gpu parallel training\u001b[39;00m\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\transformers\\trainer.py:2758\u001b[0m, in \u001b[0;36mTrainer.compute_loss\u001b[1;34m(self, model, inputs, return_outputs)\u001b[0m\n\u001b[0;32m   2756\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m   2757\u001b[0m     labels \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m-> 2758\u001b[0m outputs \u001b[38;5;241m=\u001b[39m model(\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39minputs)\n\u001b[0;32m   2759\u001b[0m \u001b[38;5;66;03m# Save past state if it exists\u001b[39;00m\n\u001b[0;32m   2760\u001b[0m \u001b[38;5;66;03m# TODO: this needs to be fixed and made cleaner later.\u001b[39;00m\n\u001b[0;32m   2761\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39margs\u001b[38;5;241m.\u001b[39mpast_index \u001b[38;5;241m>\u001b[39m\u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m:\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\torch\\nn\\modules\\module.py:1518\u001b[0m, in \u001b[0;36mModule._wrapped_call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1516\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_compiled_call_impl(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)  \u001b[38;5;66;03m# type: ignore[misc]\u001b[39;00m\n\u001b[0;32m   1517\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m-> 1518\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_call_impl(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\torch\\nn\\modules\\module.py:1527\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1522\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[0;32m   1523\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[0;32m   1524\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[0;32m   1525\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[0;32m   1526\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[1;32m-> 1527\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m forward_call(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m   1529\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m   1530\u001b[0m     result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\transformers\\models\\bert\\modeling_bert.py:1778\u001b[0m, in \u001b[0;36mBertForTokenClassification.forward\u001b[1;34m(self, input_ids, attention_mask, token_type_ids, position_ids, head_mask, inputs_embeds, labels, output_attentions, output_hidden_states, return_dict)\u001b[0m\n\u001b[0;32m   1776\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m labels \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m   1777\u001b[0m     loss_fct \u001b[38;5;241m=\u001b[39m CrossEntropyLoss()\n\u001b[1;32m-> 1778\u001b[0m     loss \u001b[38;5;241m=\u001b[39m loss_fct(logits\u001b[38;5;241m.\u001b[39mview(\u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m, \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mnum_labels), labels\u001b[38;5;241m.\u001b[39mview(\u001b[38;5;241m-\u001b[39m\u001b[38;5;241m1\u001b[39m))\n\u001b[0;32m   1780\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m return_dict:\n\u001b[0;32m   1781\u001b[0m     output \u001b[38;5;241m=\u001b[39m (logits,) \u001b[38;5;241m+\u001b[39m outputs[\u001b[38;5;241m2\u001b[39m:]\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\torch\\nn\\modules\\module.py:1518\u001b[0m, in \u001b[0;36mModule._wrapped_call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1516\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_compiled_call_impl(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)  \u001b[38;5;66;03m# type: ignore[misc]\u001b[39;00m\n\u001b[0;32m   1517\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m-> 1518\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_call_impl(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\torch\\nn\\modules\\module.py:1527\u001b[0m, in \u001b[0;36mModule._call_impl\u001b[1;34m(self, *args, **kwargs)\u001b[0m\n\u001b[0;32m   1522\u001b[0m \u001b[38;5;66;03m# If we don't have any hooks, we want to skip the rest of the logic in\u001b[39;00m\n\u001b[0;32m   1523\u001b[0m \u001b[38;5;66;03m# this function, and just call forward.\u001b[39;00m\n\u001b[0;32m   1524\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m (\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m_forward_pre_hooks\n\u001b[0;32m   1525\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_backward_pre_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_backward_hooks\n\u001b[0;32m   1526\u001b[0m         \u001b[38;5;129;01mor\u001b[39;00m _global_forward_hooks \u001b[38;5;129;01mor\u001b[39;00m _global_forward_pre_hooks):\n\u001b[1;32m-> 1527\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m forward_call(\u001b[38;5;241m*\u001b[39margs, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n\u001b[0;32m   1529\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[0;32m   1530\u001b[0m     result \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mNone\u001b[39;00m\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\torch\\nn\\modules\\loss.py:1179\u001b[0m, in \u001b[0;36mCrossEntropyLoss.forward\u001b[1;34m(self, input, target)\u001b[0m\n\u001b[0;32m   1178\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mforward\u001b[39m(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;28minput\u001b[39m: Tensor, target: Tensor) \u001b[38;5;241m-\u001b[39m\u001b[38;5;241m>\u001b[39m Tensor:\n\u001b[1;32m-> 1179\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m F\u001b[38;5;241m.\u001b[39mcross_entropy(\u001b[38;5;28minput\u001b[39m, target, weight\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mweight,\n\u001b[0;32m   1180\u001b[0m                            ignore_index\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mignore_index, reduction\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mreduction,\n\u001b[0;32m   1181\u001b[0m                            label_smoothing\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mlabel_smoothing)\n",
      "File \u001b[1;32m~\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\torch\\nn\\functional.py:3053\u001b[0m, in \u001b[0;36mcross_entropy\u001b[1;34m(input, target, weight, size_average, ignore_index, reduce, reduction, label_smoothing)\u001b[0m\n\u001b[0;32m   3051\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m size_average \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mor\u001b[39;00m reduce \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[0;32m   3052\u001b[0m     reduction \u001b[38;5;241m=\u001b[39m _Reduction\u001b[38;5;241m.\u001b[39mlegacy_get_string(size_average, reduce)\n\u001b[1;32m-> 3053\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m torch\u001b[38;5;241m.\u001b[39m_C\u001b[38;5;241m.\u001b[39m_nn\u001b[38;5;241m.\u001b[39mcross_entropy_loss(\u001b[38;5;28minput\u001b[39m, target, weight, _Reduction\u001b[38;5;241m.\u001b[39mget_enum(reduction), ignore_index, label_smoothing)\n",
      "\u001b[1;31mValueError\u001b[0m: Expected input batch_size (4400) to match target batch_size (16)."
     ]
    }
   ],
   "source": [
    "trainer.train()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0efa416d-9c57-40a5-8314-20d5ff5fb999",
   "metadata": {},
   "source": [
    "## Native Pytorch Yaklaşımı ile Eğitim"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "07dd3f46-44da-4171-8921-32f0e8d57abc",
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenized_datasets = tokenized_dataset.remove_columns([\"text\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "6b4c8762-7ef3-4a51-9d64-09a134e55406",
   "metadata": {},
   "outputs": [],
   "source": [
    "tokenized_datasets.set_format(\"torch\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2f97fe47-80eb-417c-8932-16d301c6ec34",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch.utils.data import DataLoader\n",
    "\n",
    "train_dataloader = DataLoader(tokenized_datasets['train'], shuffle=True, batch_size=8)\n",
    "eval_dataloader = DataLoader(tokenized_datasets['test'], batch_size=8) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "1e98e2bb-c0b8-41c6-a901-52581f747598",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Some weights of BertForSequenceClassification were not initialized from the model checkpoint at dbmdz/bert-base-turkish-128k-uncased and are newly initialized: ['classifier.bias', 'classifier.weight']\n",
      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n"
     ]
    }
   ],
   "source": [
    "from transformers import AutoModelForSequenceClassification\n",
    "\n",
    "model = AutoModelForSequenceClassification.from_pretrained(\"dbmdz/bert-base-turkish-128k-uncased\", num_labels=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "39b6a5e4-db59-40f4-8875-ef88a26487f7",
   "metadata": {},
   "outputs": [],
   "source": [
    "from torch.optim import AdamW\n",
    "\n",
    "optimizer = AdamW(model.parameters(), lr=5e-5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "4b4290c7-e44a-4d64-bb36-88b03065ff8f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "BertForSequenceClassification(\n",
       "  (bert): BertModel(\n",
       "    (embeddings): BertEmbeddings(\n",
       "      (word_embeddings): Embedding(128000, 768, padding_idx=0)\n",
       "      (position_embeddings): Embedding(512, 768)\n",
       "      (token_type_embeddings): Embedding(2, 768)\n",
       "      (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n",
       "      (dropout): Dropout(p=0.1, inplace=False)\n",
       "    )\n",
       "    (encoder): BertEncoder(\n",
       "      (layer): ModuleList(\n",
       "        (0-11): 12 x BertLayer(\n",
       "          (attention): BertAttention(\n",
       "            (self): BertSelfAttention(\n",
       "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
       "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
       "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
       "              (dropout): Dropout(p=0.1, inplace=False)\n",
       "            )\n",
       "            (output): BertSelfOutput(\n",
       "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
       "              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n",
       "              (dropout): Dropout(p=0.1, inplace=False)\n",
       "            )\n",
       "          )\n",
       "          (intermediate): BertIntermediate(\n",
       "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
       "            (intermediate_act_fn): GELUActivation()\n",
       "          )\n",
       "          (output): BertOutput(\n",
       "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
       "            (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n",
       "            (dropout): Dropout(p=0.1, inplace=False)\n",
       "          )\n",
       "        )\n",
       "      )\n",
       "    )\n",
       "    (pooler): BertPooler(\n",
       "      (dense): Linear(in_features=768, out_features=768, bias=True)\n",
       "      (activation): Tanh()\n",
       "    )\n",
       "  )\n",
       "  (dropout): Dropout(p=0.1, inplace=False)\n",
       "  (classifier): Linear(in_features=768, out_features=2, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from transformers import get_scheduler\n",
    "\n",
    "num_epochs = 3\n",
    "num_training_steps = num_epochs * len(train_dataloader)\n",
    "lr_scheduler = get_scheduler(\n",
    "    name=\"linear\", optimizer=optimizer, num_warmup_steps=0, num_training_steps=num_training_steps\n",
    ")\n",
    "\n",
    "model.to(device)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "4c6ac222-f1f6-4b36-ae5b-797a2850ca33",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "61172f20931048319eeb713b97a0aa81",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "  0%|          | 0/8292 [00:00<?, ?it/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from tqdm.auto import tqdm\n",
    "\n",
    "progress_bar = tqdm(range(num_training_steps))\n",
    "\n",
    "model.train()\n",
    "for epoch in range(num_epochs):\n",
    "    for batch in train_dataloader:\n",
    "        batch = {k: v.to(device) for k, v in batch.items()}\n",
    "        outputs = model(**batch)\n",
    "        loss = outputs.loss\n",
    "        loss.backward()\n",
    "\n",
    "        optimizer.step()\n",
    "        lr_scheduler.step()\n",
    "        optimizer.zero_grad()\n",
    "        progress_bar.update(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "756a5fb6-522c-41a2-98a5-92070348b429",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'accuracy': 0.9667546174142481}"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import evaluate\n",
    "\n",
    "metric = evaluate.load(\"accuracy\")\n",
    "model.eval()\n",
    "for batch in eval_dataloader:\n",
    "    batch = {k: v.to(device) for k, v in batch.items()}\n",
    "    with torch.no_grad():\n",
    "        outputs = model(**batch)\n",
    "\n",
    "    logits = outputs.logits\n",
    "    predictions = torch.argmax(logits, dim=-1)\n",
    "    metric.add_batch(predictions=predictions, references=batch[\"labels\"])\n",
    "\n",
    "metric.compute()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "df61e2c2-c1bc-4edf-b4e6-e2340006501e",
   "metadata": {},
   "source": [
    "## Eğitilen Modeli Huggingface'e Push Et"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "d6de1f4d-2da7-4f1f-b114-62dad61d5bf2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "621fe492526f4ddfb4aadceb4bae117a",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "model.safetensors:   0%|          | 0.00/737M [00:00<?, ?B/s]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "CommitInfo(commit_url='https://huggingface.co/halilibr/dilbazlar-binary-disorder-detection-model-acc-96/commit/c6b2da4e1b0477f319b585f05e7a729a74481fa0', commit_message='Upload BertForSequenceClassification', commit_description='', oid='c6b2da4e1b0477f319b585f05e7a729a74481fa0', pr_url=None, pr_revision=None, pr_num=None)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.push_to_hub('halilibr/dilbazlar-binary-disorder-detection-model-acc-96')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "bb6d7d3a-bc9f-4bc8-9ee6-9bf8e7f0b371",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CommitInfo(commit_url='https://huggingface.co/halilibr/dilbazlar-binary-disorder-detection-model-acc-96/commit/fbc675120c0c76ee4d44616cc694200376aaf9c4', commit_message='Upload tokenizer', commit_description='', oid='fbc675120c0c76ee4d44616cc694200376aaf9c4', pr_url=None, pr_revision=None, pr_num=None)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tokenizer.push_to_hub(\"halilibr/dilbazlar-binary-disorder-detection-model-acc-96\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f0e2af34-70a9-40ee-b16c-3fa58c3e4bdd",
   "metadata": {},
   "source": [
    "## Modeli Test Et"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "f15139c6-bce8-439c-8ad9-a5b7b94020b8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Text': 'Sosyal ortamlarda rahat bir şekilde iletişim kurabiliyorum.',\n",
       " 'Label': 'Normal'}"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_data[100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "ba72fbbd-27da-48ff-90a7-fc67b4923ba3",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\halilibrahim.hatun\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\huggingface_hub\\file_download.py:1150: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n",
      "  warnings.warn(\n",
      "C:\\Users\\halilibrahim.hatun\\AppData\\Local\\miniconda3\\envs\\psynexa_torch_cpu\\Lib\\site-packages\\huggingface_hub\\file_download.py:1150: FutureWarning: `resume_download` is deprecated and will be removed in version 1.0.0. Downloads always resume when possible. If you want to force a new download, use `force_download=True`.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "BertForSequenceClassification(\n",
       "  (bert): BertModel(\n",
       "    (embeddings): BertEmbeddings(\n",
       "      (word_embeddings): Embedding(128000, 768, padding_idx=0)\n",
       "      (position_embeddings): Embedding(512, 768)\n",
       "      (token_type_embeddings): Embedding(2, 768)\n",
       "      (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n",
       "      (dropout): Dropout(p=0.1, inplace=False)\n",
       "    )\n",
       "    (encoder): BertEncoder(\n",
       "      (layer): ModuleList(\n",
       "        (0-11): 12 x BertLayer(\n",
       "          (attention): BertAttention(\n",
       "            (self): BertSelfAttention(\n",
       "              (query): Linear(in_features=768, out_features=768, bias=True)\n",
       "              (key): Linear(in_features=768, out_features=768, bias=True)\n",
       "              (value): Linear(in_features=768, out_features=768, bias=True)\n",
       "              (dropout): Dropout(p=0.1, inplace=False)\n",
       "            )\n",
       "            (output): BertSelfOutput(\n",
       "              (dense): Linear(in_features=768, out_features=768, bias=True)\n",
       "              (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n",
       "              (dropout): Dropout(p=0.1, inplace=False)\n",
       "            )\n",
       "          )\n",
       "          (intermediate): BertIntermediate(\n",
       "            (dense): Linear(in_features=768, out_features=3072, bias=True)\n",
       "            (intermediate_act_fn): GELUActivation()\n",
       "          )\n",
       "          (output): BertOutput(\n",
       "            (dense): Linear(in_features=3072, out_features=768, bias=True)\n",
       "            (LayerNorm): LayerNorm((768,), eps=1e-12, elementwise_affine=True)\n",
       "            (dropout): Dropout(p=0.1, inplace=False)\n",
       "          )\n",
       "        )\n",
       "      )\n",
       "    )\n",
       "    (pooler): BertPooler(\n",
       "      (dense): Linear(in_features=768, out_features=768, bias=True)\n",
       "      (activation): Tanh()\n",
       "    )\n",
       "  )\n",
       "  (dropout): Dropout(p=0.1, inplace=False)\n",
       "  (classifier): Linear(in_features=768, out_features=2, bias=True)\n",
       ")"
      ]
     },
     "execution_count": 1,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import torch\n",
    "from transformers import AutoTokenizer, AutoModelForSequenceClassification\n",
    "\n",
    "loaded_model = AutoModelForSequenceClassification.from_pretrained(\"halilibr/dilbazlar-binary-disorder-detection-model-acc-96\", num_labels=2)\n",
    "tokenizer = AutoTokenizer.from_pretrained(\"halilibr/dilbazlar-binary-disorder-detection-model-acc-96\")\n",
    "\n",
    "# Move the model to the appropriate device\n",
    "device = torch.device(\"cuda\" if torch.cuda.is_available() else \"cpu\")\n",
    "loaded_model.to(device)\n",
    "\n",
    "# Ensure model is in evaluation mode\n",
    "loaded_model.eval()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "d08153a4-ad30-4c92-acf2-edaa24ff7163",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Example input\n",
    "input_text = \"bugün okula gittim çok güzeldi şimdi de eve geldim yine güzeldi\"\n",
    "\n",
    "def predict(input_text):\n",
    "    # Tokenize the input (ensure the tokenizer is appropriate for your model)\n",
    "    inputs = tokenizer(input_text, max_length=150, padding=\"max_length\", truncation=True, return_tensors=\"pt\")\n",
    "    \n",
    "    # Move the inputs to the appropriate device\n",
    "    inputs = {k: v.to(device) for k, v in inputs.items()}\n",
    "    \n",
    "    # Disable gradient computation for inference\n",
    "    with torch.no_grad():\n",
    "        # Forward pass to get outputs\n",
    "        outputs = loaded_model(**inputs)\n",
    "        \n",
    "        # Get the prediction\n",
    "        # Note: `AutoModel` might not include logits. Ensure you use the appropriate model class for your task.\n",
    "        if hasattr(outputs, 'logits'):\n",
    "            preds = torch.argmax(outputs.logits, dim=-1)\n",
    "        else:\n",
    "            # Handle the case where the model does not have logits (e.g., outputs are raw hidden states)\n",
    "            preds = torch.argmax(outputs[0], dim=-1)\n",
    "    \n",
    "    # Convert prediction to numpy array and print (if needed)\n",
    "    prediction = preds.cpu().numpy()[0]\n",
    "\n",
    "    return prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f272375d-eecb-45fb-8b0f-d9aaf01b7df1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n",
      "0\n"
     ]
    }
   ],
   "source": [
    "daily_phrases = [\n",
    "    \"Bugün hava çok güzel.\",\n",
    "    \"İşe gitmem gerekiyor.\",\n",
    "    \"Kahvaltı yaptın mı?\",\n",
    "    \"Sinemaya gitmek ister misin?\",\n",
    "    \"Bu hafta sonu ne yapıyorsun?\",\n",
    "    \"Yeni bir kitap okuyorum.\",\n",
    "    \"Bu yemek harika olmuş.\",\n",
    "    \"Arkadaşlarla buluşacağım.\",\n",
    "    \"Yürüyüşe çıkalım mı?\",\n",
    "    \"Yeni bir hobi edindim.\",\n",
    "    \"Bu filmi izledin mi?\",\n",
    "    \"Tatilde nereye gitmek istersin?\",\n",
    "    \"Hangi sporu yapıyorsun?\",\n",
    "    \"Ders çalışmam gerekiyor.\",\n",
    "    \"Bu hafta sonu bir parti var.\",\n",
    "    \"Alışverişe çıkmak istiyorum.\",\n",
    "    \"Spor salonuna gidiyorum.\",\n",
    "    \"Yeni bir tarif denedim.\",\n",
    "    \"Bu hafta yoğun geçecek.\",\n",
    "    \"Yeni bir projeye başladım.\",\n",
    "    \"Yarın toplantım var.\",\n",
    "    \"Bahçede çalışmak istiyorum.\",\n",
    "    \"Yeni bir müzik albümü keşfettim.\",\n",
    "    \"Bu kitabı önerir misin?\",\n",
    "    \"Yarın sabah erken kalkmalıyım.\",\n",
    "    \"Ailemi ziyaret edeceğim.\",\n",
    "    \"Yeni bir dil öğreniyorum.\",\n",
    "    \"Bu dizi çok güzel.\",\n",
    "    \"Yoga yapmaya başladım.\",\n",
    "    \"İşe erken gitmem gerekiyor.\",\n",
    "    \"Bu hafta sonu dinlenmek istiyorum.\",\n",
    "    \"Yeni bir restoran denedim.\",\n",
    "    \"Hafta sonu planın nedir?\",\n",
    "    \"Yeni bir arkadaş edindim.\",\n",
    "    \"Bu film harikaydı.\",\n",
    "    \"Bisiklet sürmeye çıkacağım.\",\n",
    "    \"Bugün spor yapmalıyım.\",\n",
    "    \"Yeni bir bilgisayar aldım.\",\n",
    "    \"Bu kitabı bitirdim.\",\n",
    "    \"Yeni bir iş görüşmem var.\",\n",
    "    \"Evin dekorasyonunu değiştirdim.\",\n",
    "    \"Bu sabah yürüyüşe çıktım.\",\n",
    "    \"Yeni bir telefon aldım.\",\n",
    "    \"Bu hafta yoğun çalışacağım.\",\n",
    "    \"Hafta sonu tatile çıkıyorum.\",\n",
    "    \"Yeni bir proje üzerinde çalışıyorum.\",\n",
    "    \"İşler yolunda gidiyor.\",\n",
    "    \"Bu kitabı sevdim.\",\n",
    "    \"Yemek yapmayı çok seviyorum.\"\n",
    "]\n",
    "\n",
    "for sentence in daily_phrases:\n",
    "    print(predict(sentence))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "e781a903-e06a-4f9f-bca6-08dd2a627441",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.10.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}