HW4_FineTuning.ipynb

{
 "metadata": {
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3",
   "language": "python"
  },
  "language_info": {
   "name": "python",
   "version": "3.10.13",
   "mimetype": "text/x-python",
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "pygments_lexer": "ipython3",
   "nbconvert_exporter": "python",
   "file_extension": ".py"
  },
  "vscode": {
   "interpreter": {
    "hash": "2d58e898dde0263bc564c6968b04150abacfd33eed9b19aaa8e45c040360e146"
   }
  },
  "kaggle": {
   "accelerator": "nvidiaTeslaT4",
   "dataSources": [
    {
     "sourceId": 80485,
     "sourceType": "modelInstanceVersion",
     "isSourceIdPinned": true,
     "modelInstanceId": 67641,
     "modelId": 92721
    }
   ],
   "dockerImageVersionId": 30747,
   "isInternetEnabled": true,
   "language": "python",
   "sourceType": "notebook",
   "isGpuEnabled": true
  }
 },
 "nbformat_minor": 4,
 "nbformat": 4,
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "# PEFT Finetuning for Neutralizing the Gender Biased Texts\n",
    "\n",
    "This notebook uses int8 quantization and LoRA finetuning.\n",
    "\n",
    "**_Note:_** To run this notebook on a machine with less than 24GB VRAM (e.g. T4 with 16GB) the context length of the training dataset needs to be adapted.\n",
    "We do this based on the available VRAM during execution.\n",
    "If you run into OOM issues try to further lower the value of train_config.context_length.\n",
    "\n",
    "We run the notebook on a machine with GPU T4 $\\times2$ accelarator."
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Step 0: Install pre-requirements and convert checkpoint\n",
    "\n",
    "We need to have llama-recipes and its dependencies installed for this notebook, as well as HuggingFace evaluation metrics. Additionally, we need to log in with the huggingface_cli and make sure that the account is able to to access the Meta Llama weights."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "! pip install --upgrade pip\n",
    "! pip install evaluate\n",
    "! pip install bert_score\n",
    "! pip install hazm\n",
    "! pip install llama-recipes ipywidgets\n",
    "\n",
    "import huggingface_hub\n",
    "huggingface_hub.login()"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T05:20:22.869098Z",
     "iopub.execute_input": "2024-07-22T05:20:22.869921Z",
     "iopub.status.idle": "2024-07-22T05:23:46.172776Z",
     "shell.execute_reply.started": "2024-07-22T05:20:22.869886Z",
     "shell.execute_reply": "2024-07-22T05:23:46.171866Z"
    },
    "trusted": true
   },
   "execution_count": 1,
   "outputs": [
    {
     "name": "stdout",
     "text": "Requirement already satisfied: pip in /opt/conda/lib/python3.10/site-packages (23.3.2)\nCollecting pip\n  Downloading pip-24.1.2-py3-none-any.whl.metadata (3.6 kB)\nDownloading pip-24.1.2-py3-none-any.whl (1.8 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m1.8/1.8 MB\u001B[0m \u001B[31m29.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0ma \u001B[36m0:00:01\u001B[0m\n\u001B[?25hInstalling collected packages: pip\n  Attempting uninstall: pip\n    Found existing installation: pip 23.3.2\n    Uninstalling pip-23.3.2:\n      Successfully uninstalled pip-23.3.2\nSuccessfully installed pip-24.1.2\nCollecting evaluate\n  Downloading evaluate-0.4.2-py3-none-any.whl.metadata (9.3 kB)\nRequirement already satisfied: datasets>=2.0.0 in /opt/conda/lib/python3.10/site-packages (from evaluate) (2.20.0)\nRequirement already satisfied: numpy>=1.17 in /opt/conda/lib/python3.10/site-packages (from evaluate) (1.26.4)\nRequirement already satisfied: dill in /opt/conda/lib/python3.10/site-packages (from evaluate) (0.3.8)\nRequirement already satisfied: pandas in /opt/conda/lib/python3.10/site-packages (from evaluate) (2.2.2)\nRequirement already satisfied: requests>=2.19.0 in /opt/conda/lib/python3.10/site-packages (from evaluate) (2.32.3)\nRequirement already satisfied: tqdm>=4.62.1 in /opt/conda/lib/python3.10/site-packages (from evaluate) (4.66.4)\nRequirement already satisfied: xxhash in /opt/conda/lib/python3.10/site-packages (from evaluate) (3.4.1)\nRequirement already satisfied: multiprocess in /opt/conda/lib/python3.10/site-packages (from evaluate) (0.70.16)\nRequirement already satisfied: fsspec>=2021.05.0 in /opt/conda/lib/python3.10/site-packages (from fsspec[http]>=2021.05.0->evaluate) (2024.5.0)\nRequirement already satisfied: huggingface-hub>=0.7.0 in /opt/conda/lib/python3.10/site-packages (from evaluate) (0.23.4)\nRequirement already satisfied: packaging in /opt/conda/lib/python3.10/site-packages (from evaluate) (21.3)\nRequirement already satisfied: filelock in /opt/conda/lib/python3.10/site-packages (from datasets>=2.0.0->evaluate) (3.13.1)\nRequirement already satisfied: pyarrow>=15.0.0 in /opt/conda/lib/python3.10/site-packages (from datasets>=2.0.0->evaluate) (16.1.0)\nRequirement already satisfied: pyarrow-hotfix in /opt/conda/lib/python3.10/site-packages (from datasets>=2.0.0->evaluate) (0.6)\nRequirement already satisfied: aiohttp in /opt/conda/lib/python3.10/site-packages (from datasets>=2.0.0->evaluate) (3.9.1)\nRequirement already satisfied: pyyaml>=5.1 in /opt/conda/lib/python3.10/site-packages (from datasets>=2.0.0->evaluate) (6.0.1)\nRequirement already satisfied: typing-extensions>=3.7.4.3 in /opt/conda/lib/python3.10/site-packages (from huggingface-hub>=0.7.0->evaluate) (4.9.0)\nRequirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /opt/conda/lib/python3.10/site-packages (from packaging->evaluate) (3.1.1)\nRequirement already satisfied: charset-normalizer<4,>=2 in /opt/conda/lib/python3.10/site-packages (from requests>=2.19.0->evaluate) (3.3.2)\nRequirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.10/site-packages (from requests>=2.19.0->evaluate) (3.6)\nRequirement already satisfied: urllib3<3,>=1.21.1 in /opt/conda/lib/python3.10/site-packages (from requests>=2.19.0->evaluate) (1.26.18)\nRequirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.10/site-packages (from requests>=2.19.0->evaluate) (2024.7.4)\nRequirement already satisfied: python-dateutil>=2.8.2 in /opt/conda/lib/python3.10/site-packages (from pandas->evaluate) (2.9.0.post0)\nRequirement already satisfied: pytz>=2020.1 in /opt/conda/lib/python3.10/site-packages (from pandas->evaluate) (2023.3.post1)\nRequirement already satisfied: tzdata>=2022.7 in /opt/conda/lib/python3.10/site-packages (from pandas->evaluate) (2023.4)\nRequirement already satisfied: attrs>=17.3.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets>=2.0.0->evaluate) (23.2.0)\nRequirement already satisfied: multidict<7.0,>=4.5 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets>=2.0.0->evaluate) (6.0.4)\nRequirement already satisfied: yarl<2.0,>=1.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets>=2.0.0->evaluate) (1.9.3)\nRequirement already satisfied: frozenlist>=1.1.1 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets>=2.0.0->evaluate) (1.4.1)\nRequirement already satisfied: aiosignal>=1.1.2 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets>=2.0.0->evaluate) (1.3.1)\nRequirement already satisfied: async-timeout<5.0,>=4.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets>=2.0.0->evaluate) (4.0.3)\nRequirement already satisfied: six>=1.5 in /opt/conda/lib/python3.10/site-packages (from python-dateutil>=2.8.2->pandas->evaluate) (1.16.0)\nDownloading evaluate-0.4.2-py3-none-any.whl (84 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m84.1/84.1 kB\u001B[0m \u001B[31m2.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hInstalling collected packages: evaluate\nSuccessfully installed evaluate-0.4.2\nCollecting bert_score\n  Downloading bert_score-0.3.13-py3-none-any.whl.metadata (15 kB)\nRequirement already satisfied: torch>=1.0.0 in /opt/conda/lib/python3.10/site-packages (from bert_score) (2.1.2)\nRequirement already satisfied: pandas>=1.0.1 in /opt/conda/lib/python3.10/site-packages (from bert_score) (2.2.2)\nRequirement already satisfied: transformers>=3.0.0 in /opt/conda/lib/python3.10/site-packages (from bert_score) (4.42.3)\nRequirement already satisfied: numpy in /opt/conda/lib/python3.10/site-packages (from bert_score) (1.26.4)\nRequirement already satisfied: requests in /opt/conda/lib/python3.10/site-packages (from bert_score) (2.32.3)\nRequirement already satisfied: tqdm>=4.31.1 in /opt/conda/lib/python3.10/site-packages (from bert_score) (4.66.4)\nRequirement already satisfied: matplotlib in /opt/conda/lib/python3.10/site-packages (from bert_score) (3.7.5)\nRequirement already satisfied: packaging>=20.9 in /opt/conda/lib/python3.10/site-packages (from bert_score) (21.3)\nRequirement already satisfied: pyparsing!=3.0.5,>=2.0.2 in /opt/conda/lib/python3.10/site-packages (from packaging>=20.9->bert_score) (3.1.1)\nRequirement already satisfied: python-dateutil>=2.8.2 in /opt/conda/lib/python3.10/site-packages (from pandas>=1.0.1->bert_score) (2.9.0.post0)\nRequirement already satisfied: pytz>=2020.1 in /opt/conda/lib/python3.10/site-packages (from pandas>=1.0.1->bert_score) (2023.3.post1)\nRequirement already satisfied: tzdata>=2022.7 in /opt/conda/lib/python3.10/site-packages (from pandas>=1.0.1->bert_score) (2023.4)\nRequirement already satisfied: filelock in /opt/conda/lib/python3.10/site-packages (from torch>=1.0.0->bert_score) (3.13.1)\nRequirement already satisfied: typing-extensions in /opt/conda/lib/python3.10/site-packages (from torch>=1.0.0->bert_score) (4.9.0)\nRequirement already satisfied: sympy in /opt/conda/lib/python3.10/site-packages (from torch>=1.0.0->bert_score) (1.13.0)\nRequirement already satisfied: networkx in /opt/conda/lib/python3.10/site-packages (from torch>=1.0.0->bert_score) (3.2.1)\nRequirement already satisfied: jinja2 in /opt/conda/lib/python3.10/site-packages (from torch>=1.0.0->bert_score) (3.1.2)\nRequirement already satisfied: fsspec in /opt/conda/lib/python3.10/site-packages (from torch>=1.0.0->bert_score) (2024.5.0)\nRequirement already satisfied: huggingface-hub<1.0,>=0.23.2 in /opt/conda/lib/python3.10/site-packages (from transformers>=3.0.0->bert_score) (0.23.4)\nRequirement already satisfied: pyyaml>=5.1 in /opt/conda/lib/python3.10/site-packages (from transformers>=3.0.0->bert_score) (6.0.1)\nRequirement already satisfied: regex!=2019.12.17 in /opt/conda/lib/python3.10/site-packages (from transformers>=3.0.0->bert_score) (2023.12.25)\nRequirement already satisfied: safetensors>=0.4.1 in /opt/conda/lib/python3.10/site-packages (from transformers>=3.0.0->bert_score) (0.4.3)\nRequirement already satisfied: tokenizers<0.20,>=0.19 in /opt/conda/lib/python3.10/site-packages (from transformers>=3.0.0->bert_score) (0.19.1)\nRequirement already satisfied: contourpy>=1.0.1 in /opt/conda/lib/python3.10/site-packages (from matplotlib->bert_score) (1.2.0)\nRequirement already satisfied: cycler>=0.10 in /opt/conda/lib/python3.10/site-packages (from matplotlib->bert_score) (0.12.1)\nRequirement already satisfied: fonttools>=4.22.0 in /opt/conda/lib/python3.10/site-packages (from matplotlib->bert_score) (4.47.0)\nRequirement already satisfied: kiwisolver>=1.0.1 in /opt/conda/lib/python3.10/site-packages (from matplotlib->bert_score) (1.4.5)\nRequirement already satisfied: pillow>=6.2.0 in /opt/conda/lib/python3.10/site-packages (from matplotlib->bert_score) (9.5.0)\nRequirement already satisfied: charset-normalizer<4,>=2 in /opt/conda/lib/python3.10/site-packages (from requests->bert_score) (3.3.2)\nRequirement already satisfied: idna<4,>=2.5 in /opt/conda/lib/python3.10/site-packages (from requests->bert_score) (3.6)\nRequirement already satisfied: urllib3<3,>=1.21.1 in /opt/conda/lib/python3.10/site-packages (from requests->bert_score) (1.26.18)\nRequirement already satisfied: certifi>=2017.4.17 in /opt/conda/lib/python3.10/site-packages (from requests->bert_score) (2024.7.4)\nRequirement already satisfied: six>=1.5 in /opt/conda/lib/python3.10/site-packages (from python-dateutil>=2.8.2->pandas>=1.0.1->bert_score) (1.16.0)\nRequirement already satisfied: MarkupSafe>=2.0 in /opt/conda/lib/python3.10/site-packages (from jinja2->torch>=1.0.0->bert_score) (2.1.3)\nRequirement already satisfied: mpmath<1.4,>=1.1.0 in /opt/conda/lib/python3.10/site-packages (from sympy->torch>=1.0.0->bert_score) (1.3.0)\nDownloading bert_score-0.3.13-py3-none-any.whl (61 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m61.1/61.1 kB\u001B[0m \u001B[31m2.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hInstalling collected packages: bert_score\nSuccessfully installed bert_score-0.3.13\nCollecting hazm\n  Downloading hazm-0.10.0-py3-none-any.whl.metadata (11 kB)\nCollecting fasttext-wheel<0.10.0,>=0.9.2 (from hazm)\n  Downloading fasttext_wheel-0.9.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (16 kB)\nRequirement already satisfied: flashtext<3.0,>=2.7 in /opt/conda/lib/python3.10/site-packages (from hazm) (2.7)\nRequirement already satisfied: gensim<5.0.0,>=4.3.1 in /opt/conda/lib/python3.10/site-packages (from hazm) (4.3.2)\nCollecting nltk<4.0.0,>=3.8.1 (from hazm)\n  Downloading nltk-3.8.1-py3-none-any.whl.metadata (2.8 kB)\nCollecting numpy==1.24.3 (from hazm)\n  Downloading numpy-1.24.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (5.6 kB)\nCollecting python-crfsuite<0.10.0,>=0.9.9 (from hazm)\n  Downloading python_crfsuite-0.9.10-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.2 kB)\nRequirement already satisfied: scikit-learn<2.0.0,>=1.2.2 in /opt/conda/lib/python3.10/site-packages (from hazm) (1.2.2)\nRequirement already satisfied: pybind11>=2.2 in /opt/conda/lib/python3.10/site-packages (from fasttext-wheel<0.10.0,>=0.9.2->hazm) (2.13.1)\nRequirement already satisfied: setuptools>=0.7.0 in /opt/conda/lib/python3.10/site-packages (from fasttext-wheel<0.10.0,>=0.9.2->hazm) (69.0.3)\nRequirement already satisfied: scipy>=1.7.0 in /opt/conda/lib/python3.10/site-packages (from gensim<5.0.0,>=4.3.1->hazm) (1.11.4)\nRequirement already satisfied: smart-open>=1.8.1 in /opt/conda/lib/python3.10/site-packages (from gensim<5.0.0,>=4.3.1->hazm) (6.4.0)\nRequirement already satisfied: click in /opt/conda/lib/python3.10/site-packages (from nltk<4.0.0,>=3.8.1->hazm) (8.1.7)\nRequirement already satisfied: joblib in /opt/conda/lib/python3.10/site-packages (from nltk<4.0.0,>=3.8.1->hazm) (1.4.2)\nRequirement already satisfied: regex>=2021.8.3 in /opt/conda/lib/python3.10/site-packages (from nltk<4.0.0,>=3.8.1->hazm) (2023.12.25)\nRequirement already satisfied: tqdm in /opt/conda/lib/python3.10/site-packages (from nltk<4.0.0,>=3.8.1->hazm) (4.66.4)\nRequirement already satisfied: threadpoolctl>=2.0.0 in /opt/conda/lib/python3.10/site-packages (from scikit-learn<2.0.0,>=1.2.2->hazm) (3.2.0)\nDownloading hazm-0.10.0-py3-none-any.whl (892 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m892.6/892.6 kB\u001B[0m \u001B[31m19.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading numpy-1.24.3-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (17.3 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m17.3/17.3 MB\u001B[0m \u001B[31m84.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading fasttext_wheel-0.9.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (4.4 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m4.4/4.4 MB\u001B[0m \u001B[31m84.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m\n\u001B[?25hDownloading nltk-3.8.1-py3-none-any.whl (1.5 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m1.5/1.5 MB\u001B[0m \u001B[31m58.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading python_crfsuite-0.9.10-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.1 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m1.1/1.1 MB\u001B[0m \u001B[31m49.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hInstalling collected packages: python-crfsuite, numpy, nltk, fasttext-wheel, hazm\n  Attempting uninstall: numpy\n    Found existing installation: numpy 1.26.4\n    Uninstalling numpy-1.26.4:\n      Successfully uninstalled numpy-1.26.4\n  Attempting uninstall: nltk\n    Found existing installation: nltk 3.2.4\n    Uninstalling nltk-3.2.4:\n      Successfully uninstalled nltk-3.2.4\n\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\ncudf 24.6.1 requires cubinlinker, which is not installed.\ncudf 24.6.1 requires cupy-cuda11x>=12.0.0, which is not installed.\ncudf 24.6.1 requires ptxcompiler, which is not installed.\ncuml 24.6.1 requires cupy-cuda11x>=12.0.0, which is not installed.\ndask-cudf 24.6.1 requires cupy-cuda11x>=12.0.0, which is not installed.\ntensorflow-decision-forests 1.8.1 requires wurlitzer, which is not installed.\nucx-py 0.38.0 requires libucx<1.16,>=1.15.0, which is not installed.\nucxx 0.38.0 requires libucx>=1.15.0, which is not installed.\nalbumentations 1.4.0 requires numpy>=1.24.4, but you have numpy 1.24.3 which is incompatible.\napache-beam 2.46.0 requires dill<0.3.2,>=0.3.1.1, but you have dill 0.3.8 which is incompatible.\napache-beam 2.46.0 requires pyarrow<10.0.0,>=3.0.0, but you have pyarrow 16.1.0 which is incompatible.\ncudf 24.6.1 requires cuda-python<12.0a0,>=11.7.1, but you have cuda-python 12.5.0 which is incompatible.\nfeaturetools 1.31.0 requires numpy>=1.25.0, but you have numpy 1.24.3 which is incompatible.\nlibpysal 4.9.2 requires packaging>=22, but you have packaging 21.3 which is incompatible.\nlibpysal 4.9.2 requires shapely>=2.0.1, but you have shapely 1.8.5.post1 which is incompatible.\nmomepy 0.7.2 requires shapely>=2, but you have shapely 1.8.5.post1 which is incompatible.\nosmnx 1.9.3 requires shapely>=2.0, but you have shapely 1.8.5.post1 which is incompatible.\npointpats 2.5.0 requires shapely>=2, but you have shapely 1.8.5.post1 which is incompatible.\npreprocessing 0.1.13 requires nltk==3.2.4, but you have nltk 3.8.1 which is incompatible.\npylibraft 24.6.0 requires cuda-python<12.0a0,>=11.7.1, but you have cuda-python 12.5.0 which is incompatible.\nrmm 24.6.0 requires cuda-python<12.0a0,>=11.7.1, but you have cuda-python 12.5.0 which is incompatible.\nspaghetti 1.7.6 requires shapely>=2.0.1, but you have shapely 1.8.5.post1 which is incompatible.\nspopt 0.6.1 requires shapely>=2.0.1, but you have shapely 1.8.5.post1 which is incompatible.\ntensorflow 2.15.0 requires keras<2.16,>=2.15.0, but you have keras 3.4.1 which is incompatible.\ntensorstore 0.1.63 requires ml-dtypes>=0.3.1, but you have ml-dtypes 0.2.0 which is incompatible.\nwoodwork 0.31.0 requires numpy>=1.25.0, but you have numpy 1.24.3 which is incompatible.\nxarray 2024.6.0 requires packaging>=23.1, but you have packaging 21.3 which is incompatible.\u001B[0m\u001B[31m\n\u001B[0mSuccessfully installed fasttext-wheel-0.9.2 hazm-0.10.0 nltk-3.8.1 numpy-1.24.3 python-crfsuite-0.9.10\nCollecting llama-recipes\n  Downloading llama_recipes-0.0.2-py3-none-any.whl.metadata (11 kB)\nRequirement already satisfied: ipywidgets in /opt/conda/lib/python3.10/site-packages (7.7.1)\nRequirement already satisfied: accelerate in /opt/conda/lib/python3.10/site-packages (from llama-recipes) (0.32.1)\nRequirement already satisfied: appdirs in /opt/conda/lib/python3.10/site-packages (from llama-recipes) (1.4.4)\nCollecting bitsandbytes (from llama-recipes)\n  Downloading bitsandbytes-0.43.1-py3-none-manylinux_2_24_x86_64.whl.metadata (2.2 kB)\nCollecting black (from llama-recipes)\n  Downloading black-24.4.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (77 kB)\n\u001B[2K     \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m77.1/77.1 kB\u001B[0m \u001B[31m3.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hCollecting chardet (from llama-recipes)\n  Downloading chardet-5.2.0-py3-none-any.whl.metadata (3.4 kB)\nRequirement already satisfied: datasets in /opt/conda/lib/python3.10/site-packages (from llama-recipes) (2.20.0)\nCollecting fire (from llama-recipes)\n  Downloading fire-0.6.0.tar.gz (88 kB)\n\u001B[2K     \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m88.4/88.4 kB\u001B[0m \u001B[31m5.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25h  Preparing metadata (setup.py) ... \u001B[?25ldone\n\u001B[?25hCollecting gradio (from llama-recipes)\n  Downloading gradio-4.38.1-py3-none-any.whl.metadata (15 kB)\nCollecting loralib (from llama-recipes)\n  Downloading loralib-0.1.2-py3-none-any.whl.metadata (15 kB)\nRequirement already satisfied: matplotlib in /opt/conda/lib/python3.10/site-packages (from llama-recipes) (3.7.5)\nCollecting openai (from llama-recipes)\n  Downloading openai-1.36.1-py3-none-any.whl.metadata (22 kB)\nCollecting optimum (from llama-recipes)\n  Downloading optimum-1.21.2-py3-none-any.whl.metadata (19 kB)\nCollecting peft (from llama-recipes)\n  Downloading peft-0.11.1-py3-none-any.whl.metadata (13 kB)\nCollecting py7zr (from llama-recipes)\n  Downloading py7zr-0.21.1-py3-none-any.whl.metadata (17 kB)\nRequirement already satisfied: scipy in /opt/conda/lib/python3.10/site-packages (from llama-recipes) (1.11.4)\nRequirement already satisfied: sentencepiece in /opt/conda/lib/python3.10/site-packages (from llama-recipes) (0.2.0)\nRequirement already satisfied: tabulate in /opt/conda/lib/python3.10/site-packages (from llama-recipes) (0.9.0)\nCollecting torch>=2.2 (from llama-recipes)\n  Downloading torch-2.3.1-cp310-cp310-manylinux1_x86_64.whl.metadata (26 kB)\nRequirement already satisfied: transformers>=4.40.0 in /opt/conda/lib/python3.10/site-packages (from llama-recipes) (4.42.3)\nCollecting typing-extensions==4.8.0 (from llama-recipes)\n  Downloading typing_extensions-4.8.0-py3-none-any.whl.metadata (3.0 kB)\nRequirement already satisfied: ipykernel>=4.5.1 in /opt/conda/lib/python3.10/site-packages (from ipywidgets) (6.28.0)\nRequirement already satisfied: ipython-genutils~=0.2.0 in /opt/conda/lib/python3.10/site-packages (from ipywidgets) (0.2.0)\nRequirement already satisfied: traitlets>=4.3.1 in /opt/conda/lib/python3.10/site-packages (from ipywidgets) (5.9.0)\nRequirement already satisfied: widgetsnbextension~=3.6.0 in /opt/conda/lib/python3.10/site-packages (from ipywidgets) (3.6.7)\nRequirement already satisfied: ipython>=4.0.0 in /opt/conda/lib/python3.10/site-packages (from ipywidgets) (8.20.0)\nRequirement already satisfied: jupyterlab-widgets>=1.0.0 in /opt/conda/lib/python3.10/site-packages (from ipywidgets) (3.0.9)\nRequirement already satisfied: comm>=0.1.1 in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (0.2.1)\nRequirement already satisfied: debugpy>=1.6.5 in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (1.8.0)\nRequirement already satisfied: jupyter-client>=6.1.12 in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (7.4.9)\nRequirement already satisfied: jupyter-core!=5.0.*,>=4.12 in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (5.7.1)\nRequirement already satisfied: matplotlib-inline>=0.1 in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (0.1.6)\nRequirement already satisfied: nest-asyncio in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (1.5.8)\nRequirement already satisfied: packaging in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (21.3)\nRequirement already satisfied: psutil in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (5.9.3)\nRequirement already satisfied: pyzmq>=24 in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (24.0.1)\nRequirement already satisfied: tornado>=6.1 in /opt/conda/lib/python3.10/site-packages (from ipykernel>=4.5.1->ipywidgets) (6.3.3)\nRequirement already satisfied: decorator in /opt/conda/lib/python3.10/site-packages (from ipython>=4.0.0->ipywidgets) (5.1.1)\nRequirement already satisfied: jedi>=0.16 in /opt/conda/lib/python3.10/site-packages (from ipython>=4.0.0->ipywidgets) (0.19.1)\nRequirement already satisfied: prompt-toolkit<3.1.0,>=3.0.41 in /opt/conda/lib/python3.10/site-packages (from ipython>=4.0.0->ipywidgets) (3.0.42)\nRequirement already satisfied: pygments>=2.4.0 in /opt/conda/lib/python3.10/site-packages (from ipython>=4.0.0->ipywidgets) (2.17.2)\nRequirement already satisfied: stack-data in /opt/conda/lib/python3.10/site-packages (from ipython>=4.0.0->ipywidgets) (0.6.2)\nRequirement already satisfied: exceptiongroup in /opt/conda/lib/python3.10/site-packages (from ipython>=4.0.0->ipywidgets) (1.2.0)\nRequirement already satisfied: pexpect>4.3 in /opt/conda/lib/python3.10/site-packages (from ipython>=4.0.0->ipywidgets) (4.8.0)\nRequirement already satisfied: filelock in /opt/conda/lib/python3.10/site-packages (from torch>=2.2->llama-recipes) (3.13.1)\nRequirement already satisfied: sympy in /opt/conda/lib/python3.10/site-packages (from torch>=2.2->llama-recipes) (1.13.0)\nRequirement already satisfied: networkx in /opt/conda/lib/python3.10/site-packages (from torch>=2.2->llama-recipes) (3.2.1)\nRequirement already satisfied: jinja2 in /opt/conda/lib/python3.10/site-packages (from torch>=2.2->llama-recipes) (3.1.2)\nRequirement already satisfied: fsspec in /opt/conda/lib/python3.10/site-packages (from torch>=2.2->llama-recipes) (2024.5.0)\nCollecting nvidia-cuda-nvrtc-cu12==12.1.105 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_cuda_nvrtc_cu12-12.1.105-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\nCollecting nvidia-cuda-runtime-cu12==12.1.105 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_cuda_runtime_cu12-12.1.105-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\nCollecting nvidia-cuda-cupti-cu12==12.1.105 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_cuda_cupti_cu12-12.1.105-py3-none-manylinux1_x86_64.whl.metadata (1.6 kB)\nCollecting nvidia-cudnn-cu12==8.9.2.26 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_cudnn_cu12-8.9.2.26-py3-none-manylinux1_x86_64.whl.metadata (1.6 kB)\nCollecting nvidia-cublas-cu12==12.1.3.1 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_cublas_cu12-12.1.3.1-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\nCollecting nvidia-cufft-cu12==11.0.2.54 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_cufft_cu12-11.0.2.54-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\nCollecting nvidia-curand-cu12==10.3.2.106 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_curand_cu12-10.3.2.106-py3-none-manylinux1_x86_64.whl.metadata (1.5 kB)\nCollecting nvidia-cusolver-cu12==11.4.5.107 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_cusolver_cu12-11.4.5.107-py3-none-manylinux1_x86_64.whl.metadata (1.6 kB)\nCollecting nvidia-cusparse-cu12==12.1.0.106 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_cusparse_cu12-12.1.0.106-py3-none-manylinux1_x86_64.whl.metadata (1.6 kB)\nCollecting nvidia-nccl-cu12==2.20.5 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_nccl_cu12-2.20.5-py3-none-manylinux2014_x86_64.whl.metadata (1.8 kB)\nCollecting nvidia-nvtx-cu12==12.1.105 (from torch>=2.2->llama-recipes)\n  Downloading nvidia_nvtx_cu12-12.1.105-py3-none-manylinux1_x86_64.whl.metadata (1.7 kB)\nCollecting triton==2.3.1 (from torch>=2.2->llama-recipes)\n  Downloading triton-2.3.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (1.4 kB)\nCollecting nvidia-nvjitlink-cu12 (from nvidia-cusolver-cu12==11.4.5.107->torch>=2.2->llama-recipes)\n  Downloading nvidia_nvjitlink_cu12-12.5.82-py3-none-manylinux2014_x86_64.whl.metadata (1.5 kB)\nRequirement already satisfied: huggingface-hub<1.0,>=0.23.2 in /opt/conda/lib/python3.10/site-packages (from transformers>=4.40.0->llama-recipes) (0.23.4)\nRequirement already satisfied: numpy<2.0,>=1.17 in /opt/conda/lib/python3.10/site-packages (from transformers>=4.40.0->llama-recipes) (1.24.3)\nRequirement already satisfied: pyyaml>=5.1 in /opt/conda/lib/python3.10/site-packages (from transformers>=4.40.0->llama-recipes) (6.0.1)\nRequirement already satisfied: regex!=2019.12.17 in /opt/conda/lib/python3.10/site-packages (from transformers>=4.40.0->llama-recipes) (2023.12.25)\nRequirement already satisfied: requests in /opt/conda/lib/python3.10/site-packages (from transformers>=4.40.0->llama-recipes) (2.32.3)\nRequirement already satisfied: safetensors>=0.4.1 in /opt/conda/lib/python3.10/site-packages (from transformers>=4.40.0->llama-recipes) (0.4.3)\nRequirement already satisfied: tokenizers<0.20,>=0.19 in /opt/conda/lib/python3.10/site-packages (from transformers>=4.40.0->llama-recipes) (0.19.1)\nRequirement already satisfied: tqdm>=4.27 in /opt/conda/lib/python3.10/site-packages (from transformers>=4.40.0->llama-recipes) (4.66.4)\nRequirement already satisfied: notebook>=4.4.1 in /opt/conda/lib/python3.10/site-packages (from widgetsnbextension~=3.6.0->ipywidgets) (6.5.4)\nRequirement already satisfied: click>=8.0.0 in /opt/conda/lib/python3.10/site-packages (from black->llama-recipes) (8.1.7)\nRequirement already satisfied: mypy-extensions>=0.4.3 in /opt/conda/lib/python3.10/site-packages (from black->llama-recipes) (1.0.0)\nCollecting packaging (from ipykernel>=4.5.1->ipywidgets)\n  Downloading packaging-24.1-py3-none-any.whl.metadata (3.2 kB)\nCollecting pathspec>=0.9.0 (from black->llama-recipes)\n  Downloading pathspec-0.12.1-py3-none-any.whl.metadata (21 kB)\nRequirement already satisfied: platformdirs>=2 in /opt/conda/lib/python3.10/site-packages (from black->llama-recipes) (3.11.0)\nRequirement already satisfied: tomli>=1.1.0 in /opt/conda/lib/python3.10/site-packages (from black->llama-recipes) (2.0.1)\nCollecting tokenize-rt>=3.2.0 (from black[jupyter]->llama-recipes)\n  Downloading tokenize_rt-5.2.0-py2.py3-none-any.whl.metadata (4.1 kB)\nRequirement already satisfied: pyarrow>=15.0.0 in /opt/conda/lib/python3.10/site-packages (from datasets->llama-recipes) (16.1.0)\nRequirement already satisfied: pyarrow-hotfix in /opt/conda/lib/python3.10/site-packages (from datasets->llama-recipes) (0.6)\nRequirement already satisfied: dill<0.3.9,>=0.3.0 in /opt/conda/lib/python3.10/site-packages (from datasets->llama-recipes) (0.3.8)\nRequirement already satisfied: pandas in /opt/conda/lib/python3.10/site-packages (from datasets->llama-recipes) (2.2.2)\nRequirement already satisfied: xxhash in /opt/conda/lib/python3.10/site-packages (from datasets->llama-recipes) (3.4.1)\nRequirement already satisfied: multiprocess in /opt/conda/lib/python3.10/site-packages (from datasets->llama-recipes) (0.70.16)\nRequirement already satisfied: aiohttp in /opt/conda/lib/python3.10/site-packages (from datasets->llama-recipes) (3.9.1)\nRequirement already satisfied: six in /opt/conda/lib/python3.10/site-packages (from fire->llama-recipes) (1.16.0)\nRequirement already satisfied: termcolor in /opt/conda/lib/python3.10/site-packages (from fire->llama-recipes) (2.4.0)\nRequirement already satisfied: aiofiles<24.0,>=22.0 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (22.1.0)\nRequirement already satisfied: altair<6.0,>=5.0 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (5.3.0)\nRequirement already satisfied: fastapi in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (0.108.0)\nCollecting ffmpy (from gradio->llama-recipes)\n  Downloading ffmpy-0.3.2.tar.gz (5.5 kB)\n  Preparing metadata (setup.py) ... \u001B[?25ldone\n\u001B[?25hCollecting gradio-client==1.1.0 (from gradio->llama-recipes)\n  Downloading gradio_client-1.1.0-py3-none-any.whl.metadata (7.1 kB)\nRequirement already satisfied: httpx>=0.24.1 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (0.27.0)\nRequirement already satisfied: importlib-resources<7.0,>=1.3 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (6.1.1)\nRequirement already satisfied: markupsafe~=2.0 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (2.1.3)\nRequirement already satisfied: orjson~=3.0 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (3.9.10)\nRequirement already satisfied: pillow<11.0,>=8.0 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (9.5.0)\nRequirement already satisfied: pydantic>=2.0 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (2.5.3)\nRequirement already satisfied: pydub in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (0.25.1)\nCollecting python-multipart>=0.0.9 (from gradio->llama-recipes)\n  Downloading python_multipart-0.0.9-py3-none-any.whl.metadata (2.5 kB)\nCollecting ruff>=0.2.2 (from gradio->llama-recipes)\n  Downloading ruff-0.5.4-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (24 kB)\nCollecting semantic-version~=2.0 (from gradio->llama-recipes)\n  Downloading semantic_version-2.10.0-py2.py3-none-any.whl.metadata (9.7 kB)\nCollecting tomlkit==0.12.0 (from gradio->llama-recipes)\n  Downloading tomlkit-0.12.0-py3-none-any.whl.metadata (2.7 kB)\nRequirement already satisfied: typer<1.0,>=0.12 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (0.12.3)\nCollecting urllib3~=2.0 (from gradio->llama-recipes)\n  Downloading urllib3-2.2.2-py3-none-any.whl.metadata (6.4 kB)\nRequirement already satisfied: uvicorn>=0.14.0 in /opt/conda/lib/python3.10/site-packages (from gradio->llama-recipes) (0.25.0)\nCollecting websockets<12.0,>=10.0 (from gradio-client==1.1.0->gradio->llama-recipes)\n  Downloading websockets-11.0.3-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.6 kB)\nRequirement already satisfied: contourpy>=1.0.1 in /opt/conda/lib/python3.10/site-packages (from matplotlib->llama-recipes) (1.2.0)\nRequirement already satisfied: cycler>=0.10 in /opt/conda/lib/python3.10/site-packages (from matplotlib->llama-recipes) (0.12.1)\nRequirement already satisfied: fonttools>=4.22.0 in /opt/conda/lib/python3.10/site-packages (from matplotlib->llama-recipes) (4.47.0)\nRequirement already satisfied: kiwisolver>=1.0.1 in /opt/conda/lib/python3.10/site-packages (from matplotlib->llama-recipes) (1.4.5)\nRequirement already satisfied: pyparsing>=2.3.1 in /opt/conda/lib/python3.10/site-packages (from matplotlib->llama-recipes) (3.1.1)\nRequirement already satisfied: python-dateutil>=2.7 in /opt/conda/lib/python3.10/site-packages (from matplotlib->llama-recipes) (2.9.0.post0)\nRequirement already satisfied: anyio<5,>=3.5.0 in /opt/conda/lib/python3.10/site-packages (from openai->llama-recipes) (4.2.0)\nRequirement already satisfied: distro<2,>=1.7.0 in /opt/conda/lib/python3.10/site-packages (from openai->llama-recipes) (1.9.0)\nRequirement already satisfied: sniffio in /opt/conda/lib/python3.10/site-packages (from openai->llama-recipes) (1.3.0)\nCollecting coloredlogs (from optimum->llama-recipes)\n  Downloading coloredlogs-15.0.1-py2.py3-none-any.whl.metadata (12 kB)\nRequirement already satisfied: texttable in /opt/conda/lib/python3.10/site-packages (from py7zr->llama-recipes) (1.7.0)\nCollecting pycryptodomex>=3.16.0 (from py7zr->llama-recipes)\n  Downloading pycryptodomex-3.20.0-cp35-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (3.4 kB)\nCollecting pyzstd>=0.15.9 (from py7zr->llama-recipes)\n  Downloading pyzstd-0.16.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (2.4 kB)\nCollecting pyppmd<1.2.0,>=1.1.0 (from py7zr->llama-recipes)\n  Downloading pyppmd-1.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (5.7 kB)\nCollecting pybcj<1.1.0,>=1.0.0 (from py7zr->llama-recipes)\n  Downloading pybcj-1.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.0 kB)\nCollecting multivolumefile>=0.2.3 (from py7zr->llama-recipes)\n  Downloading multivolumefile-0.2.3-py3-none-any.whl.metadata (6.3 kB)\nCollecting inflate64<1.1.0,>=1.0.0 (from py7zr->llama-recipes)\n  Downloading inflate64-1.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.0 kB)\nRequirement already satisfied: brotli>=1.1.0 in /opt/conda/lib/python3.10/site-packages (from py7zr->llama-recipes) (1.1.0)\nRequirement already satisfied: jsonschema>=3.0 in /opt/conda/lib/python3.10/site-packages (from altair<6.0,>=5.0->gradio->llama-recipes) (4.20.0)\nRequirement already satisfied: toolz in /opt/conda/lib/python3.10/site-packages (from altair<6.0,>=5.0->gradio->llama-recipes) (0.12.1)\nRequirement already satisfied: idna>=2.8 in /opt/conda/lib/python3.10/site-packages (from anyio<5,>=3.5.0->openai->llama-recipes) (3.6)\nRequirement already satisfied: attrs>=17.3.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->llama-recipes) (23.2.0)\nRequirement already satisfied: multidict<7.0,>=4.5 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->llama-recipes) (6.0.4)\nRequirement already satisfied: yarl<2.0,>=1.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->llama-recipes) (1.9.3)\nRequirement already satisfied: frozenlist>=1.1.1 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->llama-recipes) (1.4.1)\nRequirement already satisfied: aiosignal>=1.1.2 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->llama-recipes) (1.3.1)\nRequirement already satisfied: async-timeout<5.0,>=4.0 in /opt/conda/lib/python3.10/site-packages (from aiohttp->datasets->llama-recipes) (4.0.3)\nRequirement already satisfied: certifi in /opt/conda/lib/python3.10/site-packages (from httpx>=0.24.1->gradio->llama-recipes) (2024.7.4)\nRequirement already satisfied: httpcore==1.* in /opt/conda/lib/python3.10/site-packages (from httpx>=0.24.1->gradio->llama-recipes) (1.0.5)\nRequirement already satisfied: h11<0.15,>=0.13 in /opt/conda/lib/python3.10/site-packages (from httpcore==1.*->httpx>=0.24.1->gradio->llama-recipes) (0.14.0)\nRequirement already satisfied: parso<0.9.0,>=0.8.3 in /opt/conda/lib/python3.10/site-packages (from jedi>=0.16->ipython>=4.0.0->ipywidgets) (0.8.3)\nRequirement already satisfied: entrypoints in /opt/conda/lib/python3.10/site-packages (from jupyter-client>=6.1.12->ipykernel>=4.5.1->ipywidgets) (0.4)\nRequirement already satisfied: argon2-cffi in /opt/conda/lib/python3.10/site-packages (from notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (23.1.0)\nRequirement already satisfied: nbformat in /opt/conda/lib/python3.10/site-packages (from notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (5.9.2)\nRequirement already satisfied: nbconvert>=5 in /opt/conda/lib/python3.10/site-packages (from notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (6.4.5)\nRequirement already satisfied: Send2Trash>=1.8.0 in /opt/conda/lib/python3.10/site-packages (from notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (1.8.2)\nRequirement already satisfied: terminado>=0.8.3 in /opt/conda/lib/python3.10/site-packages (from notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.18.0)\nRequirement already satisfied: prometheus-client in /opt/conda/lib/python3.10/site-packages (from notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.19.0)\nRequirement already satisfied: nbclassic>=0.4.7 in /opt/conda/lib/python3.10/site-packages (from notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (1.0.0)\nRequirement already satisfied: pytz>=2020.1 in /opt/conda/lib/python3.10/site-packages (from pandas->datasets->llama-recipes) (2023.3.post1)\nRequirement already satisfied: tzdata>=2022.7 in /opt/conda/lib/python3.10/site-packages (from pandas->datasets->llama-recipes) (2023.4)\nRequirement already satisfied: ptyprocess>=0.5 in /opt/conda/lib/python3.10/site-packages (from pexpect>4.3->ipython>=4.0.0->ipywidgets) (0.7.0)\nRequirement already satisfied: wcwidth in /opt/conda/lib/python3.10/site-packages (from prompt-toolkit<3.1.0,>=3.0.41->ipython>=4.0.0->ipywidgets) (0.2.13)\nRequirement already satisfied: annotated-types>=0.4.0 in /opt/conda/lib/python3.10/site-packages (from pydantic>=2.0->gradio->llama-recipes) (0.6.0)\nRequirement already satisfied: pydantic-core==2.14.6 in /opt/conda/lib/python3.10/site-packages (from pydantic>=2.0->gradio->llama-recipes) (2.14.6)\nRequirement already satisfied: charset-normalizer<4,>=2 in /opt/conda/lib/python3.10/site-packages (from requests->transformers>=4.40.0->llama-recipes) (3.3.2)\nRequirement already satisfied: protobuf in /opt/conda/lib/python3.10/site-packages (from transformers[sentencepiece]<4.43.0,>=4.26.0->optimum->llama-recipes) (3.20.3)\nRequirement already satisfied: shellingham>=1.3.0 in /opt/conda/lib/python3.10/site-packages (from typer<1.0,>=0.12->gradio->llama-recipes) (1.5.4)\nRequirement already satisfied: rich>=10.11.0 in /opt/conda/lib/python3.10/site-packages (from typer<1.0,>=0.12->gradio->llama-recipes) (13.7.0)\nCollecting humanfriendly>=9.1 (from coloredlogs->optimum->llama-recipes)\n  Downloading humanfriendly-10.0-py2.py3-none-any.whl.metadata (9.2 kB)\nRequirement already satisfied: starlette<0.33.0,>=0.29.0 in /opt/conda/lib/python3.10/site-packages (from fastapi->gradio->llama-recipes) (0.32.0.post1)\nRequirement already satisfied: executing>=1.2.0 in /opt/conda/lib/python3.10/site-packages (from stack-data->ipython>=4.0.0->ipywidgets) (2.0.1)\nRequirement already satisfied: asttokens>=2.1.0 in /opt/conda/lib/python3.10/site-packages (from stack-data->ipython>=4.0.0->ipywidgets) (2.4.1)\nRequirement already satisfied: pure-eval in /opt/conda/lib/python3.10/site-packages (from stack-data->ipython>=4.0.0->ipywidgets) (0.2.2)\nRequirement already satisfied: mpmath<1.4,>=1.1.0 in /opt/conda/lib/python3.10/site-packages (from sympy->torch>=2.2->llama-recipes) (1.3.0)\nRequirement already satisfied: jsonschema-specifications>=2023.03.6 in /opt/conda/lib/python3.10/site-packages (from jsonschema>=3.0->altair<6.0,>=5.0->gradio->llama-recipes) (2023.12.1)\nRequirement already satisfied: referencing>=0.28.4 in /opt/conda/lib/python3.10/site-packages (from jsonschema>=3.0->altair<6.0,>=5.0->gradio->llama-recipes) (0.32.1)\nRequirement already satisfied: rpds-py>=0.7.1 in /opt/conda/lib/python3.10/site-packages (from jsonschema>=3.0->altair<6.0,>=5.0->gradio->llama-recipes) (0.16.2)\nRequirement already satisfied: jupyter-server>=1.8 in /opt/conda/lib/python3.10/site-packages (from nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (2.12.5)\nRequirement already satisfied: notebook-shim>=0.2.3 in /opt/conda/lib/python3.10/site-packages (from nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.2.3)\nRequirement already satisfied: mistune<2,>=0.8.1 in /opt/conda/lib/python3.10/site-packages (from nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.8.4)\nRequirement already satisfied: jupyterlab-pygments in /opt/conda/lib/python3.10/site-packages (from nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.3.0)\nRequirement already satisfied: bleach in /opt/conda/lib/python3.10/site-packages (from nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (6.1.0)\nRequirement already satisfied: pandocfilters>=1.4.1 in /opt/conda/lib/python3.10/site-packages (from nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (1.5.0)\nRequirement already satisfied: testpath in /opt/conda/lib/python3.10/site-packages (from nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.6.0)\nRequirement already satisfied: defusedxml in /opt/conda/lib/python3.10/site-packages (from nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.7.1)\nRequirement already satisfied: beautifulsoup4 in /opt/conda/lib/python3.10/site-packages (from nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (4.12.2)\nRequirement already satisfied: nbclient<0.6.0,>=0.5.0 in /opt/conda/lib/python3.10/site-packages (from nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.5.13)\nRequirement already satisfied: fastjsonschema in /opt/conda/lib/python3.10/site-packages (from nbformat->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (2.19.1)\nRequirement already satisfied: markdown-it-py>=2.2.0 in /opt/conda/lib/python3.10/site-packages (from rich>=10.11.0->typer<1.0,>=0.12->gradio->llama-recipes) (3.0.0)\nRequirement already satisfied: argon2-cffi-bindings in /opt/conda/lib/python3.10/site-packages (from argon2-cffi->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (21.2.0)\nRequirement already satisfied: jupyter-events>=0.9.0 in /opt/conda/lib/python3.10/site-packages (from jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.9.0)\nRequirement already satisfied: jupyter-server-terminals in /opt/conda/lib/python3.10/site-packages (from jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.5.1)\nRequirement already satisfied: overrides in /opt/conda/lib/python3.10/site-packages (from jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (7.4.0)\nRequirement already satisfied: websocket-client in /opt/conda/lib/python3.10/site-packages (from jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (1.7.0)\nRequirement already satisfied: mdurl~=0.1 in /opt/conda/lib/python3.10/site-packages (from markdown-it-py>=2.2.0->rich>=10.11.0->typer<1.0,>=0.12->gradio->llama-recipes) (0.1.2)\nRequirement already satisfied: cffi>=1.0.1 in /opt/conda/lib/python3.10/site-packages (from argon2-cffi-bindings->argon2-cffi->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (1.16.0)\nRequirement already satisfied: soupsieve>1.2 in /opt/conda/lib/python3.10/site-packages (from beautifulsoup4->nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (2.5)\nRequirement already satisfied: webencodings in /opt/conda/lib/python3.10/site-packages (from bleach->nbconvert>=5->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.5.1)\nRequirement already satisfied: pycparser in /opt/conda/lib/python3.10/site-packages (from cffi>=1.0.1->argon2-cffi-bindings->argon2-cffi->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (2.21)\nRequirement already satisfied: python-json-logger>=2.0.4 in /opt/conda/lib/python3.10/site-packages (from jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (2.0.7)\nRequirement already satisfied: rfc3339-validator in /opt/conda/lib/python3.10/site-packages (from jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.1.4)\nRequirement already satisfied: rfc3986-validator>=0.1.1 in /opt/conda/lib/python3.10/site-packages (from jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (0.1.1)\nRequirement already satisfied: fqdn in /opt/conda/lib/python3.10/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (1.5.1)\nRequirement already satisfied: isoduration in /opt/conda/lib/python3.10/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (20.11.0)\nRequirement already satisfied: jsonpointer>1.13 in /opt/conda/lib/python3.10/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (2.4)\nRequirement already satisfied: uri-template in /opt/conda/lib/python3.10/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (1.3.0)\nRequirement already satisfied: webcolors>=1.11 in /opt/conda/lib/python3.10/site-packages (from jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (1.13)\nRequirement already satisfied: arrow>=0.15.0 in /opt/conda/lib/python3.10/site-packages (from isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (1.3.0)\nRequirement already satisfied: types-python-dateutil>=2.8.10 in /opt/conda/lib/python3.10/site-packages (from arrow>=0.15.0->isoduration->jsonschema[format-nongpl]>=4.18.0->jupyter-events>=0.9.0->jupyter-server>=1.8->nbclassic>=0.4.7->notebook>=4.4.1->widgetsnbextension~=3.6.0->ipywidgets) (2.8.19.20240106)\nDownloading llama_recipes-0.0.2-py3-none-any.whl (64 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m64.9/64.9 kB\u001B[0m \u001B[31m2.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading typing_extensions-4.8.0-py3-none-any.whl (31 kB)\nDownloading torch-2.3.1-cp310-cp310-manylinux1_x86_64.whl (779.1 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m779.1/779.1 MB\u001B[0m \u001B[31m2.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_cublas_cu12-12.1.3.1-py3-none-manylinux1_x86_64.whl (410.6 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m410.6/410.6 MB\u001B[0m \u001B[31m1.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_cuda_cupti_cu12-12.1.105-py3-none-manylinux1_x86_64.whl (14.1 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m14.1/14.1 MB\u001B[0m \u001B[31m84.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_cuda_nvrtc_cu12-12.1.105-py3-none-manylinux1_x86_64.whl (23.7 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m23.7/23.7 MB\u001B[0m \u001B[31m70.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_cuda_runtime_cu12-12.1.105-py3-none-manylinux1_x86_64.whl (823 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m823.6/823.6 kB\u001B[0m \u001B[31m39.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading nvidia_cudnn_cu12-8.9.2.26-py3-none-manylinux1_x86_64.whl (731.7 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m731.7/731.7 MB\u001B[0m \u001B[31m1.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_cufft_cu12-11.0.2.54-py3-none-manylinux1_x86_64.whl (121.6 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m121.6/121.6 MB\u001B[0m \u001B[31m13.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_curand_cu12-10.3.2.106-py3-none-manylinux1_x86_64.whl (56.5 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m56.5/56.5 MB\u001B[0m \u001B[31m29.4 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_cusolver_cu12-11.4.5.107-py3-none-manylinux1_x86_64.whl (124.2 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m124.2/124.2 MB\u001B[0m \u001B[31m13.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_cusparse_cu12-12.1.0.106-py3-none-manylinux1_x86_64.whl (196.0 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m196.0/196.0 MB\u001B[0m \u001B[31m8.5 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_nccl_cu12-2.20.5-py3-none-manylinux2014_x86_64.whl (176.2 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m176.2/176.2 MB\u001B[0m \u001B[31m9.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading nvidia_nvtx_cu12-12.1.105-py3-none-manylinux1_x86_64.whl (99 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m99.1/99.1 kB\u001B[0m \u001B[31m995.3 kB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0ma \u001B[36m0:00:01\u001B[0m\n\u001B[?25hDownloading triton-2.3.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (168.1 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m168.1/168.1 MB\u001B[0m \u001B[31m9.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading bitsandbytes-0.43.1-py3-none-manylinux_2_24_x86_64.whl (119.8 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m119.8/119.8 MB\u001B[0m \u001B[31m13.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading black-24.4.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.8 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m1.8/1.8 MB\u001B[0m \u001B[31m60.6 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading chardet-5.2.0-py3-none-any.whl (199 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m199.4/199.4 kB\u001B[0m \u001B[31m12.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading gradio-4.38.1-py3-none-any.whl (12.4 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m12.4/12.4 MB\u001B[0m \u001B[31m89.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hDownloading gradio_client-1.1.0-py3-none-any.whl (318 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m318.1/318.1 kB\u001B[0m \u001B[31m19.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading tomlkit-0.12.0-py3-none-any.whl (37 kB)\nDownloading loralib-0.1.2-py3-none-any.whl (10 kB)\nDownloading openai-1.36.1-py3-none-any.whl (328 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m328.8/328.8 kB\u001B[0m \u001B[31m18.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading optimum-1.21.2-py3-none-any.whl (424 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m424.7/424.7 kB\u001B[0m \u001B[31m26.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading peft-0.11.1-py3-none-any.whl (251 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m251.6/251.6 kB\u001B[0m \u001B[31m16.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading py7zr-0.21.1-py3-none-any.whl (67 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m67.8/67.8 kB\u001B[0m \u001B[31m4.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading inflate64-1.0.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (93 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m93.1/93.1 kB\u001B[0m \u001B[31m6.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading multivolumefile-0.2.3-py3-none-any.whl (17 kB)\nDownloading packaging-24.1-py3-none-any.whl (53 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m54.0/54.0 kB\u001B[0m \u001B[31m3.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading pathspec-0.12.1-py3-none-any.whl (31 kB)\nDownloading pybcj-1.0.2-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (49 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m49.7/49.7 kB\u001B[0m \u001B[31m3.0 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading pycryptodomex-3.20.0-cp35-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.1 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m2.1/2.1 MB\u001B[0m \u001B[31m65.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading pyppmd-1.1.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (138 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m138.9/138.9 kB\u001B[0m \u001B[31m9.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading python_multipart-0.0.9-py3-none-any.whl (22 kB)\nDownloading pyzstd-0.16.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (413 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m413.8/413.8 kB\u001B[0m \u001B[31m20.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading ruff-0.5.4-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (10.1 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m10.1/10.1 MB\u001B[0m \u001B[31m90.9 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m0:01\u001B[0m\n\u001B[?25hDownloading semantic_version-2.10.0-py2.py3-none-any.whl (15 kB)\nDownloading tokenize_rt-5.2.0-py2.py3-none-any.whl (5.8 kB)\nDownloading urllib3-2.2.2-py3-none-any.whl (121 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m121.4/121.4 kB\u001B[0m \u001B[31m7.7 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading coloredlogs-15.0.1-py2.py3-none-any.whl (46 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m46.0/46.0 kB\u001B[0m \u001B[31m2.8 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading humanfriendly-10.0-py2.py3-none-any.whl (86 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m86.8/86.8 kB\u001B[0m \u001B[31m5.2 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading websockets-11.0.3-cp310-cp310-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (129 kB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m129.9/129.9 kB\u001B[0m \u001B[31m9.1 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m\n\u001B[?25hDownloading nvidia_nvjitlink_cu12-12.5.82-py3-none-manylinux2014_x86_64.whl (21.3 MB)\n\u001B[2K   \u001B[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001B[0m \u001B[32m21.3/21.3 MB\u001B[0m \u001B[31m73.3 MB/s\u001B[0m eta \u001B[36m0:00:00\u001B[0m:00:01\u001B[0m00:01\u001B[0m\n\u001B[?25hBuilding wheels for collected packages: fire, ffmpy\n  Building wheel for fire (setup.py) ... \u001B[?25ldone\n\u001B[?25h  Created wheel for fire: filename=fire-0.6.0-py2.py3-none-any.whl size=117031 sha256=695f4d378c4f96966f500c40a21da0fc06ec4a59743f96b66acb524aff15bbad\n  Stored in directory: /root/.cache/pip/wheels/d6/6d/5d/5b73fa0f46d01a793713f8859201361e9e581ced8c75e5c6a3\n  Building wheel for ffmpy (setup.py) ... \u001B[?25ldone\n\u001B[?25h  Created wheel for ffmpy: filename=ffmpy-0.3.2-py3-none-any.whl size=5584 sha256=0908ae3dd4b9fa9be6fe12f951b8f57e40b97858dc3fd0ff1042871946675e04\n  Stored in directory: /root/.cache/pip/wheels/bd/65/9a/671fc6dcde07d4418df0c592f8df512b26d7a0029c2a23dd81\nSuccessfully built fire ffmpy\nInstalling collected packages: ffmpy, websockets, urllib3, typing-extensions, triton, tomlkit, tokenize-rt, semantic-version, ruff, pyzstd, python-multipart, pyppmd, pycryptodomex, pybcj, pathspec, packaging, nvidia-nvtx-cu12, nvidia-nvjitlink-cu12, nvidia-nccl-cu12, nvidia-curand-cu12, nvidia-cufft-cu12, nvidia-cuda-runtime-cu12, nvidia-cuda-nvrtc-cu12, nvidia-cuda-cupti-cu12, nvidia-cublas-cu12, multivolumefile, loralib, inflate64, humanfriendly, fire, chardet, py7zr, nvidia-cusparse-cu12, nvidia-cudnn-cu12, coloredlogs, black, nvidia-cusolver-cu12, torch, openai, gradio-client, gradio, bitsandbytes, peft, optimum, llama-recipes\n  Attempting uninstall: websockets\n    Found existing installation: websockets 12.0\n    Uninstalling websockets-12.0:\n      Successfully uninstalled websockets-12.0\n  Attempting uninstall: urllib3\n    Found existing installation: urllib3 1.26.18\n    Uninstalling urllib3-1.26.18:\n      Successfully uninstalled urllib3-1.26.18\n  Attempting uninstall: typing-extensions\n    Found existing installation: typing_extensions 4.9.0\n    Uninstalling typing_extensions-4.9.0:\n      Successfully uninstalled typing_extensions-4.9.0\n  Attempting uninstall: tomlkit\n    Found existing installation: tomlkit 0.12.5\n    Uninstalling tomlkit-0.12.5:\n      Successfully uninstalled tomlkit-0.12.5\n  Attempting uninstall: packaging\n    Found existing installation: packaging 21.3\n    Uninstalling packaging-21.3:\n      Successfully uninstalled packaging-21.3\n  Attempting uninstall: torch\n    Found existing installation: torch 2.1.2\n    Uninstalling torch-2.1.2:\n      Successfully uninstalled torch-2.1.2\n\u001B[31mERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.\ncudf 24.6.1 requires cubinlinker, which is not installed.\ncudf 24.6.1 requires cupy-cuda11x>=12.0.0, which is not installed.\ncudf 24.6.1 requires ptxcompiler, which is not installed.\ncuml 24.6.1 requires cupy-cuda11x>=12.0.0, which is not installed.\ndask-cudf 24.6.1 requires cupy-cuda11x>=12.0.0, which is not installed.\nkeras-cv 0.9.0 requires keras-core, which is not installed.\ntensorflow-decision-forests 1.8.1 requires wurlitzer, which is not installed.\napache-beam 2.46.0 requires dill<0.3.2,>=0.3.1.1, but you have dill 0.3.8 which is incompatible.\napache-beam 2.46.0 requires pyarrow<10.0.0,>=3.0.0, but you have pyarrow 16.1.0 which is incompatible.\ncudf 24.6.1 requires cuda-python<12.0a0,>=11.7.1, but you have cuda-python 12.5.0 which is incompatible.\ndistributed 2024.5.1 requires dask==2024.5.1, but you have dask 2024.7.0 which is incompatible.\nfeaturetools 1.31.0 requires numpy>=1.25.0, but you have numpy 1.24.3 which is incompatible.\ngoogle-cloud-bigquery 2.34.4 requires packaging<22.0dev,>=14.3, but you have packaging 24.1 which is incompatible.\njupyterlab 4.2.3 requires jupyter-lsp>=2.0.0, but you have jupyter-lsp 1.5.1 which is incompatible.\njupyterlab-lsp 5.1.0 requires jupyter-lsp>=2.0.0, but you have jupyter-lsp 1.5.1 which is incompatible.\nkfp 2.5.0 requires google-cloud-storage<3,>=2.2.1, but you have google-cloud-storage 1.44.0 which is incompatible.\nkfp 2.5.0 requires urllib3<2.0.0, but you have urllib3 2.2.2 which is incompatible.\nlibpysal 4.9.2 requires shapely>=2.0.1, but you have shapely 1.8.5.post1 which is incompatible.\nmomepy 0.7.2 requires shapely>=2, but you have shapely 1.8.5.post1 which is incompatible.\nosmnx 1.9.3 requires shapely>=2.0, but you have shapely 1.8.5.post1 which is incompatible.\npointpats 2.5.0 requires shapely>=2, but you have shapely 1.8.5.post1 which is incompatible.\nrapids-dask-dependency 24.6.0a0 requires dask==2024.5.1, but you have dask 2024.7.0 which is incompatible.\nspaghetti 1.7.6 requires shapely>=2.0.1, but you have shapely 1.8.5.post1 which is incompatible.\nspopt 0.6.1 requires shapely>=2.0.1, but you have shapely 1.8.5.post1 which is incompatible.\ntensorflow 2.15.0 requires keras<2.16,>=2.15.0, but you have keras 3.4.1 which is incompatible.\u001B[0m\u001B[31m\n\u001B[0mSuccessfully installed bitsandbytes-0.43.1 black-24.4.2 chardet-5.2.0 coloredlogs-15.0.1 ffmpy-0.3.2 fire-0.6.0 gradio-4.38.1 gradio-client-1.1.0 humanfriendly-10.0 inflate64-1.0.0 llama-recipes-0.0.2 loralib-0.1.2 multivolumefile-0.2.3 nvidia-cublas-cu12-12.1.3.1 nvidia-cuda-cupti-cu12-12.1.105 nvidia-cuda-nvrtc-cu12-12.1.105 nvidia-cuda-runtime-cu12-12.1.105 nvidia-cudnn-cu12-8.9.2.26 nvidia-cufft-cu12-11.0.2.54 nvidia-curand-cu12-10.3.2.106 nvidia-cusolver-cu12-11.4.5.107 nvidia-cusparse-cu12-12.1.0.106 nvidia-nccl-cu12-2.20.5 nvidia-nvjitlink-cu12-12.5.82 nvidia-nvtx-cu12-12.1.105 openai-1.36.1 optimum-1.21.2 packaging-24.1 pathspec-0.12.1 peft-0.11.1 py7zr-0.21.1 pybcj-1.0.2 pycryptodomex-3.20.0 pyppmd-1.1.0 python-multipart-0.0.9 pyzstd-0.16.0 ruff-0.5.4 semantic-version-2.10.0 tokenize-rt-5.2.0 tomlkit-0.12.0 torch-2.3.1 triton-2.3.1 typing-extensions-4.8.0 urllib3-2.1.0 websockets-11.0.3\n",
     "output_type": "stream"
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "VBox(children=(HTML(value='<center> <img\\nsrc=https://huggingface.co/front/assets/huggingface_logo-noborder.sv…",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "8e3d6f40d0d84ba1912a289b88153544"
      }
     },
     "metadata": {}
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Step 1: Load the model\n",
    "\n",
    "Setup training configuration and load the model and tokenizer."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "import torch\n",
    "from transformers import LlamaForCausalLM, AutoTokenizer\n",
    "from llama_recipes.configs import train_config as TRAIN_CONFIG\n",
    "\n",
    "train_config = TRAIN_CONFIG()\n",
    "train_config.model_name = \"meta-llama/Meta-Llama-3-8B\"\n",
    "train_config.num_epochs = 3\n",
    "train_config.run_validation = False\n",
    "train_config.gradient_accumulation_steps = 4\n",
    "train_config.batch_size_training = 1\n",
    "train_config.lr = 2e-4\n",
    "train_config.use_fast_kernels = True\n",
    "train_config.use_fp16 = True\n",
    "train_config.context_length = 1024 if torch.cuda.get_device_properties(0).total_memory < 16e9 else 2048 # T4 16GB or A10 24GB\n",
    "train_config.batching_strategy = \"packing\"\n",
    "train_config.output_dir = \"meta-llama-sexbias\"\n",
    "\n",
    "from transformers import BitsAndBytesConfig\n",
    "config = BitsAndBytesConfig(\n",
    "    load_in_8bit=True,\n",
    ")\n",
    "\n",
    "model = LlamaForCausalLM.from_pretrained(\n",
    "            train_config.model_name,\n",
    "            device_map=\"auto\",\n",
    "            quantization_config=config,\n",
    "            use_cache=False,\n",
    "            attn_implementation=\"sdpa\" if train_config.use_fast_kernels else None,\n",
    "            torch_dtype=torch.float16,\n",
    "        )\n",
    "\n",
    "tokenizer = AutoTokenizer.from_pretrained(train_config.model_name)\n",
    "tokenizer.pad_token = tokenizer.eos_token"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T05:24:00.685872Z",
     "iopub.execute_input": "2024-07-22T05:24:00.686222Z",
     "iopub.status.idle": "2024-07-22T05:25:54.687241Z",
     "shell.execute_reply.started": "2024-07-22T05:24:00.686186Z",
     "shell.execute_reply": "2024-07-22T05:25:54.686268Z"
    },
    "trusted": true
   },
   "execution_count": 2,
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "config.json:   0%|          | 0.00/654 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "584e60425000478c8f7f309403df2679"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model.safetensors.index.json:   0%|          | 0.00/23.9k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "0a2405647e954e0cba9037048b6c398b"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading shards:   0%|          | 0/4 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "ebe2d5faa88b4c45bd7a3a6ef11af10f"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model-00001-of-00004.safetensors:   0%|          | 0.00/4.98G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "fb77d8e5d45540f68f08d96f3845d5f3"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model-00002-of-00004.safetensors:   0%|          | 0.00/5.00G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "519800a6252b413cb589f5bf1fd1c4b9"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model-00003-of-00004.safetensors:   0%|          | 0.00/4.92G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "89b2a526b7554ea990e6874231602bf0"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model-00004-of-00004.safetensors:   0%|          | 0.00/1.17G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "93e516efd1174e4eba20288145836b35"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Loading checkpoint shards:   0%|          | 0/4 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "569e83fbeee14bb880251e3ba2d073c5"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "generation_config.json:   0%|          | 0.00/177 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "5465374bbc394321bd0fca5f9386a3f0"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "tokenizer_config.json:   0%|          | 0.00/50.6k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "185647351f384d3b92aa94b5b5f53fc4"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "tokenizer.json:   0%|          | 0.00/9.09M [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "fb5bf68b67364b57901bc5d4350b5a2f"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "special_tokens_map.json:   0%|          | 0.00/73.0 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "2e5901d1667a4cae8192ff9f0f3f283b"
      }
     },
     "metadata": {}
    },
    {
     "name": "stderr",
     "text": "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Step 2: Test the Base Model\n",
    "\n",
    "Here, we test the base model of `Llama3-8b` on the test set, by the evalation metrics of BLEU and BERTScore."
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Prepare the Test Dataset"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "import datasets\n",
    "\n",
    "# Load the test dataset\n",
    "test_dataset = datasets.load_dataset(\"AmirMohammadFakhimi/gender_neutralize\", split=\"test\")\n",
    "\n",
    "# Prepare the prompts for evaluation\n",
    "prompt_template = (\n",
    "        f\"این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\\n{{dialog}}\\n________\\nمتن بدون بایاس جنسیتی:\\n\"\n",
    "    )\n",
    "\n",
    "def prepare_prompt(sample):\n",
    "    return {\n",
    "        \"prompt\": prompt_template.format(dialog=sample[\"dialogue\"]),\n",
    "        \"summary\": sample[\"summary\"]\n",
    "    }\n",
    "\n",
    "test_dataset = test_dataset.map(prepare_prompt, remove_columns=list(test_dataset.features))"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T03:58:11.907413Z",
     "iopub.execute_input": "2024-07-22T03:58:11.907911Z",
     "iopub.status.idle": "2024-07-22T03:58:15.655936Z",
     "shell.execute_reply.started": "2024-07-22T03:58:11.907882Z",
     "shell.execute_reply": "2024-07-22T03:58:15.654985Z"
    },
    "trusted": true
   },
   "execution_count": 3,
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading readme:   0%|          | 0.00/24.0 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "e010c0e7ddf248de9d129ab437b7238e"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading data:   0%|          | 0.00/332k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "ef011fac6ed14d0c9103ce78b355be44"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading data:   0%|          | 0.00/39.1k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "f0bc0f577ef444478f3a5292e08c6e31"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Generating train split:   0%|          | 0/1309 [00:00<?, ? examples/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "54dff31816964b19b513ce9ea18ded5f"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Generating test split:   0%|          | 0/146 [00:00<?, ? examples/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "0de60040f43e4de692e3b295eace2bd7"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Map:   0%|          | 0/146 [00:00<?, ? examples/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "68f5201e08e64296a865dcffd4ffdd86"
      }
     },
     "metadata": {}
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Generate Predictions"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "def generate_predictions(model, tokenizer, dataset):\n",
    "    predictions = []\n",
    "    references = []\n",
    "\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        for sample in dataset:\n",
    "            model_input = tokenizer(sample[\"prompt\"], return_tensors=\"pt\").to(\"cuda\")\n",
    "            output = model.generate(**model_input, max_new_tokens=100)\n",
    "            prediction = tokenizer.decode(output[0], skip_special_tokens=True)\n",
    "        \n",
    "            # Extract the part after \"متن بدون بایاس جنسیتی:\"\n",
    "            prediction = prediction.split('\\n________\\nمتن بدون بایاس جنسیتی:\\n')[1].strip()\n",
    "            prediction = re.sub('متن بدون بایاس جنسیتی:', ' '  , prediction)\n",
    "            # Remove non-Alphabetical characters and digits execpt dot\n",
    "            prediction = re.sub(r'[^.آ-یA-Za-z\\s]', ' ', prediction)\n",
    "            # Define a regular expression pattern that matches one or more spaces\n",
    "            pattern = re.compile(r\" +\")\n",
    "            # Apply the pattern to the text and replace the matches with a single space\n",
    "            prediction = pattern.sub(\" \", prediction)\n",
    "            \n",
    "            predictions.append(prediction)\n",
    "            references.append(sample[\"summary\"])\n",
    "\n",
    "    return predictions, references\n",
    "\n",
    "# Generate predictions for base model\n",
    "base_predictions, references = generate_predictions(model, tokenizer, test_dataset)"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-21T23:31:29.230035Z",
     "iopub.execute_input": "2024-07-21T23:31:29.230701Z",
     "iopub.status.idle": "2024-07-22T00:05:41.388738Z",
     "shell.execute_reply.started": "2024-07-21T23:31:29.230668Z",
     "shell.execute_reply": "2024-07-22T00:05:41.387906Z"
    },
    "trusted": true
   },
   "execution_count": 8,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Compute the Evaluation Metrics"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "from evaluate import load\n",
    "from hazm import word_tokenize\n",
    "\n",
    "bertscore_metric = load(\"bertscore\")\n",
    "bleu_metric = load('bleu')\n",
    "\n",
    "def compute_metrics(predictions, references):\n",
    "    # fix the references with the BLEU score HuggingFace format\n",
    "    references = [[ref] for ref in references]\n",
    "\n",
    "    # Compute BLEU score\n",
    "    bleu_score = bleu_metric.compute(predictions=predictions, references=references, tokenizer=word_tokenize)#[\"bleu\"]\n",
    "\n",
    "    # Compute BERTScore\n",
    "    bertscore = bertscore_metric.compute(predictions=predictions, references=references, lang=\"fa\", model_type='microsoft/deberta-v2-xxlarge-mnli')\n",
    "    bertscore_f1 = sum(bertscore[\"f1\"]) / len(bertscore[\"f1\"])\n",
    "\n",
    "    return bleu_score, bertscore_f1\n",
    "\n",
    "# Compute metrics for both models\n",
    "base_bleu, base_bertscore = compute_metrics(base_predictions, references)\n",
    "\n",
    "print(f\"Base Model - BLEU: {base_bleu}, BERTScore F1: {base_bertscore}\")"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T00:06:34.524362Z",
     "iopub.execute_input": "2024-07-22T00:06:34.525074Z",
     "iopub.status.idle": "2024-07-22T00:08:31.192074Z",
     "shell.execute_reply.started": "2024-07-22T00:06:34.525035Z",
     "shell.execute_reply": "2024-07-22T00:08:31.191088Z"
    },
    "trusted": true
   },
   "execution_count": 9,
   "outputs": [
    {
     "name": "stderr",
     "text": "/opt/conda/lib/python3.10/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: '/opt/conda/lib/python3.10/site-packages/torchvision/image.so: undefined symbol: _ZN3c1017RegisterOperatorsD1Ev'If you don't plan on using image functionality from `torchvision.io`, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have `libjpeg` or `libpng` installed before building `torchvision` from source?\n  warn(\n",
     "output_type": "stream"
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading builder script:   0%|          | 0.00/7.95k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "b7e7c3aab96d4dcb8113f08ba069fec1"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading builder script:   0%|          | 0.00/5.94k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "eef299836ed94eb7bd31659e505cb1bf"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading extra modules:   0%|          | 0.00/1.55k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "a341ad0eaab542c1990d93e95ffbf69f"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading extra modules:   0%|          | 0.00/3.34k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "4bbe3579e53f4f16bc21a0a1ce0bf223"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "tokenizer_config.json:   0%|          | 0.00/52.0 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "918ff2946189463db9fa62dfa2d02bc3"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "config.json:   0%|          | 0.00/952 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "20cd81c209e24162b2b2eb73b623bc37"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "spm.model:   0%|          | 0.00/2.45M [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "f598174277144696b23ab8a888d92427"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "pytorch_model.bin:   0%|          | 0.00/3.13G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "3485e088289442b2a122f2326c5e1a81"
      }
     },
     "metadata": {}
    },
    {
     "name": "stdout",
     "text": "Base Model - BLEU: {'bleu': 0.19212972085620741, 'precisions': [0.2688223938223938, 0.2023511755877939, 0.17159916926272067, 0.145979492714517], 'brevity_penalty': 1.0, 'length_ratio': 2.5009052504526252, 'translation_length': 4144, 'reference_length': 1657}, BERTScore F1: 0.818024289730477\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Sample Predictions\n",
    "Run the base model on some example inputs:"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "مردان باید در برنامه‌ریزی برای تعطیلات خانوادگی پیش‌قدم باشند.\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))\n",
    "    \n",
    "torch.cuda.empty_cache()"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T05:26:23.242761Z",
     "iopub.execute_input": "2024-07-22T05:26:23.243301Z",
     "iopub.status.idle": "2024-07-22T05:26:43.790938Z",
     "shell.execute_reply.started": "2024-07-22T05:26:23.243268Z",
     "shell.execute_reply": "2024-07-22T05:26:43.790047Z"
    },
    "trusted": true
   },
   "execution_count": 3,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n2024-07-22 05:26:26.752843: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n2024-07-22 05:26:26.752940: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n2024-07-22 05:26:26.906626: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nمردان باید در برنامه‌ریزی برای تعطیلات خانوادگی پیش‌قدم باشند.\n________\nمتن بدون بایاس جنسیتی:\nشما باید در برنامه‌ریزی برای تعطیلات خانوادگی پیش‌قدم باشید.\n\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "افراد ممکن است به جنبه های مختلف زندگی خود توجه کنند و تمرکز متفاوتی بر روی مهارت های حرفه ای داشته باشند.\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))\n",
    "    \n",
    "torch.cuda.empty_cache()"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T05:27:06.770271Z",
     "iopub.execute_input": "2024-07-22T05:27:06.771300Z",
     "iopub.status.idle": "2024-07-22T05:27:36.907211Z",
     "shell.execute_reply.started": "2024-07-22T05:27:06.771262Z",
     "shell.execute_reply": "2024-07-22T05:27:36.906252Z"
    },
    "trusted": true
   },
   "execution_count": 4,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nافراد ممکن است به جنبه های مختلف زندگی خود توجه کنند و تمرکز متفاوتی بر روی مهارت های حرفه ای داشته باشند.\n________\nمتن بدون بایاس جنسیتی:\nافراد ممکن است به جنبه های مختلف زندگی خود توجه کنند و تمرکز متفاوتی بر روی مهارت های حرفه ای داشته باشند.\n________\nمتن بدون بایاس جنسیتی:\nافراد ممکن است به جنبه های مختلف زندگی خود توجه کنند و تمرکز متفاوتی بر روی مهارت های حرفه ای داشته باشند.\n________\nمتن بدون بایاس جنسیتی:\nافراد ممکن است به جنبه های مختلف زندگی خود توجه کنند و تمرکز متفاوتی بر روی مه\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "زنان باید سرپرست سازمان‌دهی رویدادهای اجتماعی باشند.\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))\n",
    "    \n",
    "torch.cuda.empty_cache()"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T05:27:42.358569Z",
     "iopub.execute_input": "2024-07-22T05:27:42.358947Z",
     "iopub.status.idle": "2024-07-22T05:27:48.090076Z",
     "shell.execute_reply.started": "2024-07-22T05:27:42.358914Z",
     "shell.execute_reply": "2024-07-22T05:27:48.089189Z"
    },
    "trusted": true
   },
   "execution_count": 5,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nزنان باید سرپرست سازمان‌دهی رویدادهای اجتماعی باشند.\n________\nمتن بدون بایاس جنسیتی:\nزنان و مردان باید سرپرست سازمان‌دهی رویدادهای اجتماعی باشند.\n\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "المیرا اسم یک دختر است\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-21T23:28:34.924984Z",
     "iopub.execute_input": "2024-07-21T23:28:34.925637Z",
     "iopub.status.idle": "2024-07-21T23:29:04.514666Z",
     "shell.execute_reply.started": "2024-07-21T23:28:34.925603Z",
     "shell.execute_reply": "2024-07-21T23:29:04.513657Z"
    },
    "trusted": true
   },
   "execution_count": 5,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nالمیرا اسم یک دختر است\n________\nمتن بدون بایاس جنسیتی:\nالمیرا اسم یک دختر است\n________\nمتن با بایاس جنسیتی:\nالمیرا اسم یک دختر است\n________\nمتن بدون بایاس جنسیتی:\nالمیرا اسم یک دختر است\n________\nمتن با بایاس جنسیتی:\nالمیرا اسم یک دختر است\n________\nمتن بدون بایاس جنسیتی:\nالمیرا اسم یک دختر است\n________\nمتن با بایاس جنسیتی:\nالمیرا اسم یک دختر است\n________\n\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "زن‌ها نمی‌توانند به خوبی مردها طراحی کنند.\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T05:29:48.249554Z",
     "iopub.execute_input": "2024-07-22T05:29:48.250400Z",
     "iopub.status.idle": "2024-07-22T05:30:16.989779Z",
     "shell.execute_reply.started": "2024-07-22T05:29:48.250366Z",
     "shell.execute_reply": "2024-07-22T05:30:16.988779Z"
    },
    "trusted": true
   },
   "execution_count": 8,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nزن‌ها نمی‌توانند به خوبی مردها طراحی کنند.\n________\nمتن بدون بایاس جنسیتی:\nزن‌ها می‌توانند به خوبی مردها طراحی کنند.\n\nمتن با بایاس جنسیتی:\nزن‌ها نمی‌توانند به خوبی مردها طراحی کنند.\n\nمتن با بایاس جنسیتی:\nزن‌ها نمی‌توانند به خوبی مردها طراحی کنند.\n\nمتن با بایاس جنسیتی:\nزن‌ها نمی‌توانند به خوبی مردها طراحی کنند.\n\nمتن با بایاس جنسیتی:\nزن‌ها نمی‌توانند به خوبی مردها طراحی کنند.\n\nمتن با\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "زن‌ها در ورزش‌های سنگین موفق نمی‌شوند\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T05:31:16.473831Z",
     "iopub.execute_input": "2024-07-22T05:31:16.474627Z",
     "iopub.status.idle": "2024-07-22T05:31:19.906204Z",
     "shell.execute_reply.started": "2024-07-22T05:31:16.474592Z",
     "shell.execute_reply": "2024-07-22T05:31:19.905241Z"
    },
    "trusted": true
   },
   "execution_count": 9,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nزن‌ها در ورزش‌های سنگین موفق نمی‌شوند\n________\nمتن بدون بایاس جنسیتی:\nزن‌ها در ورزش‌های سنگین موفق نمی‌شوند\n\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "As you can see, the base model outputs are incorrect and sometimes, are invalid also; and get repeated weirdly.\n",
    "\n",
    "### Step 3: Load the preprocessed dataset\n",
    "\n",
    "We load and preprocess the samsum dataset which consists of curated pairs of dialogs and their summarization:"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "from dataclasses import dataclass\n",
    " \n",
    "@dataclass\n",
    "class samsum_dataset:\n",
    "    dataset: str =  \"samsum_dataset\"\n",
    "    train_split: str = \"train\"\n",
    "    test_split: str = \"validation\""
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T03:58:31.193157Z",
     "iopub.execute_input": "2024-07-22T03:58:31.193559Z",
     "iopub.status.idle": "2024-07-22T03:58:31.199288Z",
     "shell.execute_reply.started": "2024-07-22T03:58:31.193528Z",
     "shell.execute_reply": "2024-07-22T03:58:31.198203Z"
    },
    "trusted": true
   },
   "execution_count": 4,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [
    "# Copyright (c) Meta Platforms, Inc. and affiliates.\n",
    "# This software may be used and distributed according to the terms of the Llama 3 Community License Agreement.\n",
    "\n",
    "# For dataset details visit: https://huggingface.co/datasets/AmirMohammadFakhimi/gender_neutralize\n",
    "\n",
    "import copy\n",
    "import datasets\n",
    "\n",
    "def get_preprocessed_samsum(dataset_config, tokenizer, split):\n",
    "    dataset = datasets.load_dataset(\"AmirMohammadFakhimi/gender_neutralize\", split=split)\n",
    "\n",
    "    prompt = (\n",
    "        f\"این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\\n{{dialog}}\\n________\\nمتن بدون بایاس جنسیتی:\\n\"\n",
    "    )\n",
    "\n",
    "    def apply_prompt_template(sample):\n",
    "        return {\n",
    "            \"prompt\": prompt.format(dialog=sample[\"dialogue\"]),\n",
    "            \"summary\": sample[\"summary\"],\n",
    "        }\n",
    "\n",
    "    dataset = dataset.map(apply_prompt_template, remove_columns=list(dataset.features))\n",
    "\n",
    "    def tokenize_add_label(sample):\n",
    "        prompt = tokenizer.encode(tokenizer.bos_token + sample[\"prompt\"], add_special_tokens=False)\n",
    "        summary = tokenizer.encode(sample[\"summary\"] +  tokenizer.eos_token, add_special_tokens=False)\n",
    "\n",
    "        sample = {\n",
    "            \"input_ids\": prompt + summary,\n",
    "            \"attention_mask\" : [1] * (len(prompt) + len(summary)),\n",
    "            \"labels\": [-100] * len(prompt) + summary,\n",
    "            }\n",
    "\n",
    "        return sample\n",
    "\n",
    "    dataset = dataset.map(tokenize_add_label, remove_columns=list(dataset.features))\n",
    "\n",
    "    return dataset"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T03:58:31.833718Z",
     "iopub.execute_input": "2024-07-22T03:58:31.834079Z",
     "iopub.status.idle": "2024-07-22T03:58:31.843160Z",
     "shell.execute_reply.started": "2024-07-22T03:58:31.834051Z",
     "shell.execute_reply": "2024-07-22T03:58:31.842115Z"
    },
    "trusted": true
   },
   "execution_count": 5,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [
    "def get_preprocessed_dataset(\n",
    "    tokenizer, dataset_config, split: str = \"train\"\n",
    ") -> torch.utils.data.Dataset:\n",
    "\n",
    "    def get_split():\n",
    "        return (\n",
    "            dataset_config.train_split\n",
    "            if split == \"train\"\n",
    "            else dataset_config.test_split\n",
    "        )\n",
    "\n",
    "    return get_preprocessed_samsum(\n",
    "        dataset_config,\n",
    "        tokenizer,\n",
    "        get_split(),\n",
    "    )"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T03:58:33.505673Z",
     "iopub.execute_input": "2024-07-22T03:58:33.506013Z",
     "iopub.status.idle": "2024-07-22T03:58:33.511493Z",
     "shell.execute_reply.started": "2024-07-22T03:58:33.505988Z",
     "shell.execute_reply": "2024-07-22T03:58:33.510481Z"
    },
    "trusted": true
   },
   "execution_count": 6,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [
    "from llama_recipes.data.concatenator import ConcatDataset\n",
    "from llama_recipes.utils.config_utils import get_dataloader_kwargs\n",
    "\n",
    "train_dataset = get_preprocessed_dataset(tokenizer, samsum_dataset, 'train')\n",
    "\n",
    "train_dl_kwargs = get_dataloader_kwargs(train_config, train_dataset, tokenizer, \"train\")\n",
    "\n",
    "if train_config.batching_strategy == \"packing\":\n",
    "    train_dataset = ConcatDataset(train_dataset, chunk_size=train_config.context_length)\n",
    "\n",
    "train_dataloader = torch.utils.data.DataLoader(\n",
    "    train_dataset,\n",
    "    num_workers=train_config.num_workers_dataloader,\n",
    "    pin_memory=True,\n",
    "    **train_dl_kwargs,\n",
    ")"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T03:58:34.584382Z",
     "iopub.execute_input": "2024-07-22T03:58:34.584719Z",
     "iopub.status.idle": "2024-07-22T03:58:49.021722Z",
     "shell.execute_reply.started": "2024-07-22T03:58:34.584695Z",
     "shell.execute_reply": "2024-07-22T03:58:49.020775Z"
    },
    "trusted": true
   },
   "execution_count": 7,
   "outputs": [
    {
     "name": "stderr",
     "text": "2024-07-22 03:58:38.292629: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n2024-07-22 03:58:38.292732: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n2024-07-22 03:58:38.418632: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
     "output_type": "stream"
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Map:   0%|          | 0/1309 [00:00<?, ? examples/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "a69fd1c5f827447c8d3200df0f67404e"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Map:   0%|          | 0/1309 [00:00<?, ? examples/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "49811a8142294ee484eeca79ca35ff70"
      }
     },
     "metadata": {}
    },
    {
     "name": "stderr",
     "text": "Preprocessing dataset: 100%|██████████| 1309/1309 [00:00<00:00, 5204.91it/s]\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Step 4: Prepare model for PEFT\n",
    "\n",
    "Let's prepare the model for Parameter Efficient Fine Tuning (PEFT):"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "from peft import get_peft_model, prepare_model_for_kbit_training, LoraConfig\n",
    "from dataclasses import asdict\n",
    "from llama_recipes.configs import lora_config as LORA_CONFIG\n",
    "\n",
    "lora_config = LORA_CONFIG()\n",
    "lora_config.r = 8\n",
    "lora_config.lora_alpha = 32\n",
    "lora_dropout: float=0.01\n",
    "\n",
    "peft_config = LoraConfig(**asdict(lora_config))\n",
    "\n",
    "model = prepare_model_for_kbit_training(model)\n",
    "model = get_peft_model(model, peft_config)"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T03:58:52.933261Z",
     "iopub.execute_input": "2024-07-22T03:58:52.933663Z",
     "iopub.status.idle": "2024-07-22T03:58:53.138740Z",
     "shell.execute_reply.started": "2024-07-22T03:58:52.933632Z",
     "shell.execute_reply": "2024-07-22T03:58:53.137905Z"
    },
    "trusted": true
   },
   "execution_count": 8,
   "outputs": []
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Step 5: Fine tune the model\n",
    "\n",
    "Here, we fine tune the model for a single epoch."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "import torch.optim as optim\n",
    "from llama_recipes.utils.train_utils import train\n",
    "from torch.optim.lr_scheduler import StepLR\n",
    "\n",
    "model.train()\n",
    "\n",
    "optimizer = optim.AdamW(\n",
    "            model.parameters(),\n",
    "            lr=train_config.lr,\n",
    "            weight_decay=train_config.weight_decay,\n",
    "        )\n",
    "scheduler = StepLR(optimizer, step_size=1, gamma=train_config.gamma)\n",
    "\n",
    "# Start the training process\n",
    "results = train(\n",
    "    model,\n",
    "    train_dataloader,\n",
    "    None,\n",
    "    tokenizer,\n",
    "    optimizer,\n",
    "    scheduler,\n",
    "    train_config.gradient_accumulation_steps,\n",
    "    train_config,\n",
    "    None,\n",
    "    None,\n",
    "    None,\n",
    "    wandb_run=None,\n",
    ")"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T03:58:58.065118Z",
     "iopub.execute_input": "2024-07-22T03:58:58.065500Z",
     "iopub.status.idle": "2024-07-22T04:28:50.103014Z",
     "shell.execute_reply.started": "2024-07-22T03:58:58.065469Z",
     "shell.execute_reply": "2024-07-22T04:28:50.102006Z"
    },
    "trusted": true
   },
   "execution_count": 9,
   "outputs": [
    {
     "name": "stderr",
     "text": "/opt/conda/lib/python3.10/site-packages/torch/cuda/memory.py:330: FutureWarning: torch.cuda.reset_max_memory_allocated now calls torch.cuda.reset_peak_memory_stats, which resets /all/ peak memory stats.\n  warnings.warn(\nTraining Epoch: 1:   0%|\u001B[34m          \u001B[0m| 0/21 [00:00<?, ?it/s]/opt/conda/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.\n  self.pid = os.fork()\nhuggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\nTo 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/opt/conda/lib/python3.10/site-packages/torch/utils/checkpoint.py:464: UserWarning: torch.utils.checkpoint: the use_reentrant parameter should be passed explicitly. In version 2.4 we will raise an exception if use_reentrant is not passed. use_reentrant=False is recommended, but if you need to preserve the current default behavior, you can pass use_reentrant=True. Refer to docs for more details on the differences between the two variants.\n  warnings.warn(\n/opt/conda/lib/python3.10/site-packages/bitsandbytes/autograd/_functions.py:316: UserWarning: MatMul8bitLt: inputs will be cast from torch.float32 to float16 during quantization\n  warnings.warn(f\"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization\")\nTraining Epoch: 1/3, step 83/84 completed (loss: 0.02069789357483387): 100%|\u001B[34m██████████\u001B[0m| 21/21 [09:57<00:00, 28.50s/it]  /opt/conda/lib/python3.10/multiprocessing/popen_fork.py:66: RuntimeWarning: os.fork() was called. os.fork() is incompatible with multithreaded code, and JAX is multithreaded, so this will likely lead to a deadlock.\n  self.pid = os.fork()\nTraining Epoch: 1/3, step 83/84 completed (loss: 0.02069789357483387): 100%|\u001B[34m██████████\u001B[0m| 21/21 [09:57<00:00, 28.45s/it]\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "Max CUDA memory allocated was 9 GB\nMax CUDA memory reserved was 9 GB\nPeak active CUDA memory was 9 GB\nCUDA Malloc retries : 0\nCPU Total Peak Memory consumed during the train (max): 3 GB\nEpoch 1: train_perplexity=1.1242, train_epoch_loss=0.1171, epoch time 598.1484485689999s\n",
     "output_type": "stream"
    },
    {
     "name": "stderr",
     "text": "Training Epoch: 2:   0%|\u001B[34m          \u001B[0m| 0/21 [00:00<?, ?it/s]huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\nTo 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)\nTraining Epoch: 2/3, step 83/84 completed (loss: 0.017545850947499275): 100%|\u001B[34m██████████\u001B[0m| 21/21 [09:56<00:00, 28.41s/it]  \n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "Max CUDA memory allocated was 9 GB\nMax CUDA memory reserved was 9 GB\nPeak active CUDA memory was 9 GB\nCUDA Malloc retries : 0\nCPU Total Peak Memory consumed during the train (max): 4 GB\nEpoch 2: train_perplexity=1.0731, train_epoch_loss=0.0705, epoch time 597.295847465s\n",
     "output_type": "stream"
    },
    {
     "name": "stderr",
     "text": "Training Epoch: 3:   0%|\u001B[34m          \u001B[0m| 0/21 [00:00<?, ?it/s]huggingface/tokenizers: The current process just got forked, after parallelism has already been used. Disabling parallelism to avoid deadlocks...\nTo 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)\nTraining Epoch: 3/3, step 83/84 completed (loss: 0.015038029290735722): 100%|\u001B[34m██████████\u001B[0m| 21/21 [09:55<00:00, 28.37s/it]  \n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "Max CUDA memory allocated was 9 GB\nMax CUDA memory reserved was 9 GB\nPeak active CUDA memory was 9 GB\nCUDA Malloc retries : 0\nCPU Total Peak Memory consumed during the train (max): 4 GB\nEpoch 3: train_perplexity=1.0572, train_epoch_loss=0.0556, epoch time 596.5726425649998s\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Step 6:\n",
    "Save model checkpoint"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "model.save_pretrained(train_config.output_dir)"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T01:48:16.976180Z",
     "iopub.execute_input": "2024-07-22T01:48:16.977161Z",
     "iopub.status.idle": "2024-07-22T01:48:17.104786Z",
     "shell.execute_reply.started": "2024-07-22T01:48:16.977126Z",
     "shell.execute_reply": "2024-07-22T01:48:17.103951Z"
    },
    "trusted": true
   },
   "execution_count": 11,
   "outputs": []
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Load the fine-tuned model (We recommend don't run this cell untill you get CUDA OutOfMemory Error)"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "from transformers import LlamaForCausalLM, AutoTokenizer, BitsAndBytesConfig\n",
    "from peft import PeftModel, PeftConfig\n",
    "import torch\n",
    "\n",
    "# Load the base model and tokenizer\n",
    "base_model_name = \"meta-llama/Meta-Llama-3-8B\"\n",
    "tokenizer = AutoTokenizer.from_pretrained(base_model_name)\n",
    "tokenizer.pad_token = tokenizer.eos_token\n",
    "\n",
    "# Use mixed precision and quantization to save memory\n",
    "config = BitsAndBytesConfig(\n",
    "    load_in_8bit=True,\n",
    "    llm_int8_enable_fp32_cpu_offload=True  # Enable CPU offload to save GPU memory\n",
    ")\n",
    "\n",
    "base_model = LlamaForCausalLM.from_pretrained(\n",
    "    base_model_name,\n",
    "    device_map=\"auto\",\n",
    "    quantization_config=config,\n",
    "    torch_dtype=torch.float16\n",
    ")\n",
    "\n",
    "# Load the adapter configuration\n",
    "peft_model_id = '/kaggle/input/meta-llama-sexbias/pytorch/default/1'\n",
    "adapter_config = PeftConfig.from_pretrained(peft_model_id)\n",
    "\n",
    "# Load the LoRA adapter\n",
    "model = PeftModel.from_pretrained(base_model, peft_model_id, config=adapter_config)"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T02:58:58.000362Z",
     "iopub.execute_input": "2024-07-22T02:58:58.000683Z",
     "iopub.status.idle": "2024-07-22T03:01:57.405400Z",
     "shell.execute_reply.started": "2024-07-22T02:58:58.000659Z",
     "shell.execute_reply": "2024-07-22T03:01:57.404525Z"
    },
    "trusted": true
   },
   "execution_count": 2,
   "outputs": [
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "tokenizer_config.json:   0%|          | 0.00/50.6k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "5f42d395688e4784819f2405e7918707"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "tokenizer.json:   0%|          | 0.00/9.09M [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "090aafa66c574347883825714f622ecf"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "special_tokens_map.json:   0%|          | 0.00/73.0 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "6d3f52f6739b47998ed09498a41ebd91"
      }
     },
     "metadata": {}
    },
    {
     "name": "stderr",
     "text": "Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.\n",
     "output_type": "stream"
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "config.json:   0%|          | 0.00/654 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "926314b3d3724e04a2271f6b8d3b4b18"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model.safetensors.index.json:   0%|          | 0.00/23.9k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "dbd10f7a7c244828b42a4d7b10d0fee2"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading shards:   0%|          | 0/4 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "a630361633da481cbec0653d15eb85c7"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model-00001-of-00004.safetensors:   0%|          | 0.00/4.98G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "bdb9d1d0debe4f7cbbcf32436e95cf49"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model-00002-of-00004.safetensors:   0%|          | 0.00/5.00G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "34e2911d806b4fc7a15f2c05d0ff18ac"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model-00003-of-00004.safetensors:   0%|          | 0.00/4.92G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "08b3580c1cce4b2582797eb5fb6ab489"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "model-00004-of-00004.safetensors:   0%|          | 0.00/1.17G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "c0e7b5f91d3b4f5bac8ba52c004f9169"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Loading checkpoint shards:   0%|          | 0/4 [00:00<?, ?it/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "2f127c0860164c64a9c01059c9f69fa6"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "generation_config.json:   0%|          | 0.00/177 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "d6e26d3521964412bc48c3faeb5c001e"
      }
     },
     "metadata": {}
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "### Step 7: Test the Fine-tuned Model\n",
    "\n",
    "Here, we test the base model of `Llama3-8b-fineTuned` on the test set, by the evalation metrics of BLEU and BERTScore."
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### generate Predictions"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "import re\n",
    "\n",
    "def generate_predictions(model, tokenizer, dataset):\n",
    "    predictions = []\n",
    "    references = []\n",
    "\n",
    "    model.eval()\n",
    "    with torch.no_grad():\n",
    "        for sample in dataset:\n",
    "            model_input = tokenizer(sample[\"prompt\"], return_tensors=\"pt\").to(\"cuda\")\n",
    "            output = model.generate(**model_input, max_new_tokens=100)\n",
    "            prediction = tokenizer.decode(output[0], skip_special_tokens=True)\n",
    "            \n",
    "            # Extract the part after \"متن بدون بایاس جنسیتی:\"\n",
    "            prediction = prediction.split('\\n________\\nمتن بدون بایاس جنسیتی:\\n')[1].strip()\n",
    "            prediction = re.sub('متن بدون بایاس جنسیتی:', ' '  , prediction)\n",
    "            # Remove non-Alphabetical characters and digits execpt dot\n",
    "            prediction = re.sub(r'[^.آ-یA-Za-z\\s]', ' ', prediction)\n",
    "            # Define a regular expression pattern that matches one or more spaces\n",
    "            pattern = re.compile(r\" +\")\n",
    "            # Apply the pattern to the text and replace the matches with a single space\n",
    "            prediction = pattern.sub(\" \", prediction)\n",
    "            \n",
    "            predictions.append(prediction)\n",
    "            references.append(sample[\"summary\"])\n",
    "            \n",
    "            # Clear CUDA cache to free up memory\n",
    "            torch.cuda.empty_cache()\n",
    "\n",
    "    return predictions, references\n",
    "\n",
    "# Generate predictions for base model\n",
    "tuned_predictions, references = generate_predictions(model, tokenizer, test_dataset)"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T04:30:59.468383Z",
     "iopub.execute_input": "2024-07-22T04:30:59.469260Z",
     "iopub.status.idle": "2024-07-22T04:47:05.932519Z",
     "shell.execute_reply.started": "2024-07-22T04:30:59.469225Z",
     "shell.execute_reply": "2024-07-22T04:47:05.931697Z"
    },
    "trusted": true
   },
   "execution_count": 10,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\nSetting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "Save the predictions and references for test set."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "with open(\"tuned_predictions.txt\", \"w\") as f:\n",
    "    for pred in tuned_predictions:\n",
    "        f.write(str(pred) +\"\\n\")\n",
    "        \n",
    "with open(\"references.txt\", \"w\") as f:\n",
    "    for ref in references:\n",
    "        f.write(str(ref) +\"\\n\")"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T04:48:01.105867Z",
     "iopub.execute_input": "2024-07-22T04:48:01.106234Z",
     "iopub.status.idle": "2024-07-22T04:48:01.112560Z",
     "shell.execute_reply.started": "2024-07-22T04:48:01.106206Z",
     "shell.execute_reply": "2024-07-22T04:48:01.111709Z"
    },
    "trusted": true
   },
   "execution_count": 11,
   "outputs": []
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Compute the Evaluation Metrics"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "Load the predictions and references for test set.\n",
    "\n",
    "**_Note:_** Don't run this cell untill you faced CUDA OutOfMemory Error."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "references = []\n",
    "with open(\"references.txt\", \"r\") as f:\n",
    "  for line in f:\n",
    "    references.append(int(line.strip()))\n",
    "    \n",
    "tuned_predictions = []\n",
    "with open(\"tuned_predictions.txt\", \"r\") as f:\n",
    "  for line in f:\n",
    "    tuned_predictions.append(int(line.strip()))"
   ],
   "metadata": {},
   "execution_count": null,
   "outputs": []
  },
  {
   "cell_type": "code",
   "source": [
    "from evaluate import load\n",
    "from hazm import word_tokenize\n",
    "\n",
    "bertscore_metric = load(\"bertscore\")\n",
    "bleu_metric = load('bleu')\n",
    "\n",
    "def compute_metrics(predictions, references):\n",
    "    # fix the references with the BLEU score HuggingFace format\n",
    "    references = [[ref] for ref in references]\n",
    "\n",
    "    # Compute BLEU score\n",
    "    bleu_score = bleu_metric.compute(predictions=predictions, references=references, tokenizer=word_tokenize)#[\"bleu\"]\n",
    "\n",
    "    # Compute BERTScore\n",
    "    bertscore = bertscore_metric.compute(predictions=predictions, references=references, lang=\"fa\", model_type='microsoft/deberta-v2-xxlarge-mnli')\n",
    "    bertscore_f1 = sum(bertscore[\"f1\"]) / len(bertscore[\"f1\"])\n",
    "\n",
    "    return bleu_score, bertscore_f1\n",
    "\n",
    "# Compute metrics for fine-tuned model\n",
    "tuned_bleu, tuned_bertscore = compute_metrics(tuned_predictions, references)\n",
    "\n",
    "print(f\"Fine-tuned Model - BLEU: {tuned_bleu}, BERTScore F1: {tuned_bertscore}\")"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T05:02:41.132724Z",
     "iopub.execute_input": "2024-07-22T05:02:41.133086Z",
     "iopub.status.idle": "2024-07-22T05:03:36.865370Z",
     "shell.execute_reply.started": "2024-07-22T05:02:41.133061Z",
     "shell.execute_reply": "2024-07-22T05:03:36.864403Z"
    },
    "trusted": true
   },
   "execution_count": 19,
   "outputs": [
    {
     "name": "stderr",
     "text": "/opt/conda/lib/python3.10/site-packages/torchvision/io/image.py:13: UserWarning: Failed to load image Python extension: '/opt/conda/lib/python3.10/site-packages/torchvision/image.so: undefined symbol: _ZN3c1017RegisterOperatorsD1Ev'If you don't plan on using image functionality from `torchvision.io`, you can ignore this warning. Otherwise, there might be something wrong with your environment. Did you have `libjpeg` or `libpng` installed before building `torchvision` from source?\n  warn(\n",
     "output_type": "stream"
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading builder script:   0%|          | 0.00/7.95k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "8ccbda6e66df4a46ac1a4e013a837d8d"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading builder script:   0%|          | 0.00/5.94k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "87ab8216aa5f48a8a69786fc69348701"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading extra modules:   0%|          | 0.00/1.55k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "cfe5ad0bd3cb4f648b1f6d69f153e962"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "Downloading extra modules:   0%|          | 0.00/3.34k [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "b3bdbb692c614f20bc1b6a3163fd8a80"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "tokenizer_config.json:   0%|          | 0.00/52.0 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "1aa682c72e504594ac5eb0796d62eaf2"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "config.json:   0%|          | 0.00/952 [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "a7de2601a9d74446984103f68dc838b5"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "spm.model:   0%|          | 0.00/2.45M [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "63916137325b41f4bce87721ea726c71"
      }
     },
     "metadata": {}
    },
    {
     "output_type": "display_data",
     "data": {
      "text/plain": "pytorch_model.bin:   0%|          | 0.00/3.13G [00:00<?, ?B/s]",
      "application/vnd.jupyter.widget-view+json": {
       "version_major": 2,
       "version_minor": 0,
       "model_id": "5105b502aadb4785b7fb222f54bf806b"
      }
     },
     "metadata": {}
    },
    {
     "name": "stdout",
     "text": "Fine-tuned Model - BLEU: {'bleu': 0.4282493906811996, 'precisions': [0.6205770277626566, 0.46599645180366644, 0.37928802588996763, 0.30664760543245173], 'brevity_penalty': 1.0, 'length_ratio': 1.1086300543150271, 'translation_length': 1837, 'reference_length': 1657}, BERTScore F1: 0.9234269032739613\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "markdown",
   "source": [
    "#### Sample Predictions\n",
    "Run the fine-tuned model on some example inputs:"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "زن‌ها در ورزش‌های سنگین موفق نمی‌شوند\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T04:50:32.447649Z",
     "iopub.execute_input": "2024-07-22T04:50:32.448037Z",
     "iopub.status.idle": "2024-07-22T04:50:40.425280Z",
     "shell.execute_reply.started": "2024-07-22T04:50:32.448007Z",
     "shell.execute_reply": "2024-07-22T04:50:40.424416Z"
    },
    "trusted": true
   },
   "execution_count": 12,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nزن‌ها در ورزش‌های سنگین موفق نمی‌شوند\n________\nمتن بدون بایاس جنسیتی:\nموفقیت در ورزش‌های سنگین به توانایی‌های فیزیکی و ذهنی فرد بستگی دارد\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "المیرا اسم یک دختر است\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T04:51:07.288927Z",
     "iopub.execute_input": "2024-07-22T04:51:07.289893Z",
     "iopub.status.idle": "2024-07-22T04:51:09.979323Z",
     "shell.execute_reply.started": "2024-07-22T04:51:07.289855Z",
     "shell.execute_reply": "2024-07-22T04:51:09.978281Z"
    },
    "trusted": true
   },
   "execution_count": 14,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nالمیرا اسم یک دختر است\n________\nمتن بدون بایاس جنسیتی:\nالمیرا اسم یک دختر است\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "زن‌ها نمی‌توانند به خوبی مردها طراحی کنند.\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T04:51:40.752849Z",
     "iopub.execute_input": "2024-07-22T04:51:40.753209Z",
     "iopub.status.idle": "2024-07-22T04:51:44.375597Z",
     "shell.execute_reply.started": "2024-07-22T04:51:40.753180Z",
     "shell.execute_reply": "2024-07-22T04:51:44.374621Z"
    },
    "trusted": true
   },
   "execution_count": 15,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nزن‌ها نمی‌توانند به خوبی مردها طراحی کنند.\n________\nمتن بدون بایاس جنسیتی:\nهرکسی می‌تواند به خوبی طراحی کند.\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "زنان باید سرپرست سازمان‌دهی رویدادهای اجتماعی باشند.\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T04:54:46.115156Z",
     "iopub.execute_input": "2024-07-22T04:54:46.115562Z",
     "iopub.status.idle": "2024-07-22T04:54:56.534985Z",
     "shell.execute_reply.started": "2024-07-22T04:54:46.115532Z",
     "shell.execute_reply": "2024-07-22T04:54:56.534075Z"
    },
    "trusted": true
   },
   "execution_count": 16,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nزنان باید سرپرست سازمان‌دهی رویدادهای اجتماعی باشند.\n________\nمتن بدون بایاس جنسیتی:\nسرپرستی سازماندهی رویدادهای اجتماعی می‌تواند بر اساس توانایی‌های فردی و علاقه‌های شخصی تقسیم شود.\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "افراد ممکن است به جنبه های مختلف زندگی خود توجه کنند و تمرکز متفاوتی بر روی مهارت های حرفه ای داشته باشند.\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T04:57:51.987512Z",
     "iopub.execute_input": "2024-07-22T04:57:51.987917Z",
     "iopub.status.idle": "2024-07-22T04:58:02.275693Z",
     "shell.execute_reply.started": "2024-07-22T04:57:51.987889Z",
     "shell.execute_reply": "2024-07-22T04:58:02.274646Z"
    },
    "trusted": true
   },
   "execution_count": 17,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nافراد ممکن است به جنبه های مختلف زندگی خود توجه کنند و تمرکز متفاوتی بر روی مهارت های حرفه ای داشته باشند.\n________\nمتن بدون بایاس جنسیتی:\nافراد ممکن است به جنبه های مختلف زندگی خود توجه کنند و تمرکز متفاوتی بر روی مهارت های حرفه ای داشته باشند.\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [
    "eval_prompt = \"\"\"\n",
    "این متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\n",
    "مردان باید در برنامه‌ریزی برای تعطیلات خانوادگی پیش‌قدم باشند.\n",
    "________\n",
    "متن بدون بایاس جنسیتی:\n",
    "\"\"\"\n",
    "\n",
    "model_input = tokenizer(eval_prompt, return_tensors=\"pt\").to(\"cuda\")\n",
    "\n",
    "model.eval()\n",
    "with torch.no_grad():\n",
    "    print(tokenizer.decode(model.generate(**model_input, max_new_tokens=100)[0], skip_special_tokens=True))\n",
    "    \n",
    "torch.cuda.empty_cache()"
   ],
   "metadata": {
    "execution": {
     "iopub.status.busy": "2024-07-22T05:00:48.986738Z",
     "iopub.execute_input": "2024-07-22T05:00:48.987497Z",
     "iopub.status.idle": "2024-07-22T05:00:59.263072Z",
     "shell.execute_reply.started": "2024-07-22T05:00:48.987465Z",
     "shell.execute_reply": "2024-07-22T05:00:59.262069Z"
    },
    "trusted": true
   },
   "execution_count": 18,
   "outputs": [
    {
     "name": "stderr",
     "text": "Setting `pad_token_id` to `eos_token_id`:128001 for open-end generation.\n",
     "output_type": "stream"
    },
    {
     "name": "stdout",
     "text": "\nاین متن را بدون بایاس جنسیتی کن و بایاس جنسیتی آنرا از بین ببر:\nمردان باید در برنامه‌ریزی برای تعطیلات خانوادگی پیش‌قدم باشند.\n________\nمتن بدون بایاس جنسیتی:\nبرنامه‌ریزی برای تعطیلات خانوادگی باید به عهده‌ی فردی باشد که توانایی‌های لازم را دارد.\n",
     "output_type": "stream"
    }
   ]
  },
  {
   "cell_type": "code",
   "source": [],
   "metadata": {},
   "execution_count": null,
   "outputs": []
  }
 ]
}