Fine_tune_Llama_2.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU",
    "widgets": {
      "application/vnd.jupyter.widget-state+json": {
        "ad3aefa473b24f14ab44d5ab3067cc1a": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_09bb3b4001514eed8638b2094ee6d903",
              "IPY_MODEL_41db7902d4414b978838c78c624c942e",
              "IPY_MODEL_bac866fb454c42b68f1118fdc0b3b305"
            ],
            "layout": "IPY_MODEL_c3e033aafe614a1c938a076c9783631e"
          }
        },
        "09bb3b4001514eed8638b2094ee6d903": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b6d00ffc2a184a0c86ec46a5f45db26b",
            "placeholder": "​",
            "style": "IPY_MODEL_b9f4081551a8463b95df7c3ee802a45d",
            "value": "Loading checkpoint shards: 100%"
          }
        },
        "41db7902d4414b978838c78c624c942e": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_b83529889ab8421184cc0368ca23de5b",
            "max": 2,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_3b746c608411451eb577ad73543656ac",
            "value": 2
          }
        },
        "bac866fb454c42b68f1118fdc0b3b305": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e7d66667449f4c65bb9c6b7795d9f4dd",
            "placeholder": "​",
            "style": "IPY_MODEL_6b2ddab8b63d4b69838463af41aa450b",
            "value": " 2/2 [00:55&lt;00:00, 25.65s/it]"
          }
        },
        "c3e033aafe614a1c938a076c9783631e": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b6d00ffc2a184a0c86ec46a5f45db26b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "b9f4081551a8463b95df7c3ee802a45d": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "b83529889ab8421184cc0368ca23de5b": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3b746c608411451eb577ad73543656ac": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "e7d66667449f4c65bb9c6b7795d9f4dd": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "6b2ddab8b63d4b69838463af41aa450b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "cff4c007189341db853f12e91ab926b3": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HBoxModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HBoxModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HBoxView",
            "box_style": "",
            "children": [
              "IPY_MODEL_8d57720cb7dd41618193249f3c3a164b",
              "IPY_MODEL_cdca0bbac7ac4c1096f46f4bd0f8e2b6",
              "IPY_MODEL_031089c370934c57b9e631c7db9b247c"
            ],
            "layout": "IPY_MODEL_ef6f7ae8e41643a0bb372a0869ba059d"
          }
        },
        "8d57720cb7dd41618193249f3c3a164b": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_523c5f99629d4a258f1f53b63b32c029",
            "placeholder": "​",
            "style": "IPY_MODEL_3138e4baaaf24cdeacbd09888dd56383",
            "value": "Loading checkpoint shards: 100%"
          }
        },
        "cdca0bbac7ac4c1096f46f4bd0f8e2b6": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "FloatProgressModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "FloatProgressModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "ProgressView",
            "bar_style": "success",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_e51e970e57774d589110ea29bf2c258f",
            "max": 2,
            "min": 0,
            "orientation": "horizontal",
            "style": "IPY_MODEL_f76695423cdb419d80e883c2a34ae1c9",
            "value": 2
          }
        },
        "031089c370934c57b9e631c7db9b247c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "HTMLModel",
          "model_module_version": "1.5.0",
          "state": {
            "_dom_classes": [],
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "HTMLModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/controls",
            "_view_module_version": "1.5.0",
            "_view_name": "HTMLView",
            "description": "",
            "description_tooltip": null,
            "layout": "IPY_MODEL_6dda331a137943a59f2e86273f9a4317",
            "placeholder": "​",
            "style": "IPY_MODEL_eb9e401d0cf948509f5b29a804e73b4c",
            "value": " 2/2 [01:00&lt;00:00, 28.11s/it]"
          }
        },
        "ef6f7ae8e41643a0bb372a0869ba059d": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "523c5f99629d4a258f1f53b63b32c029": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "3138e4baaaf24cdeacbd09888dd56383": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        },
        "e51e970e57774d589110ea29bf2c258f": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "f76695423cdb419d80e883c2a34ae1c9": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "ProgressStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "ProgressStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "bar_color": null,
            "description_width": ""
          }
        },
        "6dda331a137943a59f2e86273f9a4317": {
          "model_module": "@jupyter-widgets/base",
          "model_name": "LayoutModel",
          "model_module_version": "1.2.0",
          "state": {
            "_model_module": "@jupyter-widgets/base",
            "_model_module_version": "1.2.0",
            "_model_name": "LayoutModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "LayoutView",
            "align_content": null,
            "align_items": null,
            "align_self": null,
            "border": null,
            "bottom": null,
            "display": null,
            "flex": null,
            "flex_flow": null,
            "grid_area": null,
            "grid_auto_columns": null,
            "grid_auto_flow": null,
            "grid_auto_rows": null,
            "grid_column": null,
            "grid_gap": null,
            "grid_row": null,
            "grid_template_areas": null,
            "grid_template_columns": null,
            "grid_template_rows": null,
            "height": null,
            "justify_content": null,
            "justify_items": null,
            "left": null,
            "margin": null,
            "max_height": null,
            "max_width": null,
            "min_height": null,
            "min_width": null,
            "object_fit": null,
            "object_position": null,
            "order": null,
            "overflow": null,
            "overflow_x": null,
            "overflow_y": null,
            "padding": null,
            "right": null,
            "top": null,
            "visibility": null,
            "width": null
          }
        },
        "eb9e401d0cf948509f5b29a804e73b4c": {
          "model_module": "@jupyter-widgets/controls",
          "model_name": "DescriptionStyleModel",
          "model_module_version": "1.5.0",
          "state": {
            "_model_module": "@jupyter-widgets/controls",
            "_model_module_version": "1.5.0",
            "_model_name": "DescriptionStyleModel",
            "_view_count": null,
            "_view_module": "@jupyter-widgets/base",
            "_view_module_version": "1.2.0",
            "_view_name": "StyleView",
            "description_width": ""
          }
        }
      }
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "#**Fine-Tune Llama 2**\n"
      ],
      "metadata": {
        "id": "OSHlAbqzDFDq"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Gerekli paketler**"
      ],
      "metadata": {
        "id": "q57Zf8p9ivZa"
      }
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "metadata": {
        "id": "GLXwJqbjtPho"
      },
      "outputs": [],
      "source": [
        "!pip install -q accelerate==0.21.0 peft==0.4.0 bitsandbytes==0.40.2 transformers==4.31.0 trl==0.4.7"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "import torch\n",
        "from datasets import load_dataset\n",
        "from transformers import (\n",
        "    AutoModelForCausalLM,\n",
        "    AutoTokenizer,\n",
        "    BitsAndBytesConfig,\n",
        "    HfArgumentParser,\n",
        "    TrainingArguments,\n",
        "    pipeline,\n",
        "    logging,\n",
        ")\n",
        "from peft import LoraConfig, PeftModel\n",
        "from trl import SFTTrainer"
      ],
      "metadata": {
        "id": "nAMzy_0FtaUZ"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Llama 2'de chat için kullanılan modellerde aşağıdaki template kullanılır.**"
      ],
      "metadata": {
        "id": "J5d58CfLoWZj"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "System Prompt (opsiyonel) : modele rehberlik etmekete.\n",
        "\n",
        "\n",
        "User prompt (gerekli) : kullanıcı talimatlarını içerir.\n",
        "\n",
        "\n",
        "Model Answer (gerekli)"
      ],
      "metadata": {
        "id": "5lYYzhCRov5y"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "![image.png](data:image/png;base64,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)"
      ],
      "metadata": {
        "id": "AS1Ee8JunXgp"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Llama 2 templateine uygun data : https://huggingface.co/datasets/mertbozkurt/llama2-TR-recipe"
      ],
      "metadata": {
        "id": "ck708D51o-h3"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Eğer Llama 2 modelinin chat verisyonu yerine ana versiyonunu kullnılacaksa spesifik prompt' gerek yok."
      ],
      "metadata": {
        "id": "jSFfMkSpphFS"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Llama 2 Fine-tune**"
      ],
      "metadata": {
        "id": "zdcR5JHdp6qY"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Full fine-tuning düşük ram veya ekran kartı imkanlarıyla mümkün değildir. Bu yüzden parameter-efficient fine-tuning (PEFT) tekniklerini kullanmak gerekir.( LoRA/QLoRA)"
      ],
      "metadata": {
        "id": "DIeTamzXqcC3"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "En iyi şekilde ihtiyaçları düşürmek için QLoRA ile 4-bit precision kullancağız."
      ],
      "metadata": {
        "id": "1lDdMCfyqvy2"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "* QLoRA rank :64\n",
        "* scaling parameter: 16.\n",
        "* Llama 2 model  4-bit precision, NF4 type\n",
        "* one epoch"
      ],
      "metadata": {
        "id": "l6CPejEgv7hq"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# The model that you want to train from the Hugging Face hub\n",
        "model_name = \"NousResearch/Llama-2-7b-chat-hf\"\n",
        "\n",
        "# The instruction dataset to use\n",
        "dataset_name = \"mertbozkurt/llama2-TR-recipe\"\n",
        "\n",
        "# Fine-tuned model name\n",
        "new_model = \"Llama-2-7b-chat-finetune\"\n",
        "\n",
        "################################################################################\n",
        "# QLoRA parameters\n",
        "################################################################################\n",
        "\n",
        "# LoRA attention dimension\n",
        "lora_r = 64\n",
        "\n",
        "# Alpha parameter for LoRA scaling\n",
        "lora_alpha = 16\n",
        "\n",
        "# Dropout probability for LoRA layers\n",
        "lora_dropout = 0.1\n",
        "\n",
        "################################################################################\n",
        "# bitsandbytes parameters\n",
        "################################################################################\n",
        "\n",
        "# Activate 4-bit precision base model loading\n",
        "use_4bit = True\n",
        "\n",
        "# Compute dtype for 4-bit base models\n",
        "bnb_4bit_compute_dtype = \"float16\"\n",
        "\n",
        "# Quantization type (fp4 or nf4)\n",
        "bnb_4bit_quant_type = \"nf4\"\n",
        "\n",
        "# Activate nested quantization for 4-bit base models (double quantization)\n",
        "use_nested_quant = False\n",
        "\n",
        "################################################################################\n",
        "# TrainingArguments parameters\n",
        "################################################################################\n",
        "\n",
        "# Output directory where the model predictions and checkpoints will be stored\n",
        "output_dir = \"./results\"\n",
        "\n",
        "# Number of training epochs\n",
        "num_train_epochs = 1\n",
        "\n",
        "# Enable fp16/bf16 training (set bf16 to True with an A100)\n",
        "fp16 = False\n",
        "bf16 = False\n",
        "\n",
        "# Batch size per GPU for training\n",
        "per_device_train_batch_size = 4\n",
        "\n",
        "# Batch size per GPU for evaluation\n",
        "per_device_eval_batch_size = 4\n",
        "\n",
        "# Number of update steps to accumulate the gradients for\n",
        "gradient_accumulation_steps = 1\n",
        "\n",
        "# Enable gradient checkpointing\n",
        "gradient_checkpointing = True\n",
        "\n",
        "# Maximum gradient normal (gradient clipping)\n",
        "max_grad_norm = 0.3\n",
        "\n",
        "# Initial learning rate (AdamW optimizer)\n",
        "learning_rate = 2e-4\n",
        "\n",
        "# Weight decay to apply to all layers except bias/LayerNorm weights\n",
        "weight_decay = 0.001\n",
        "\n",
        "# Optimizer to use\n",
        "optim = \"paged_adamw_32bit\"\n",
        "\n",
        "# Learning rate schedule\n",
        "lr_scheduler_type = \"cosine\"\n",
        "\n",
        "# Number of training steps (overrides num_train_epochs)\n",
        "max_steps = -1\n",
        "\n",
        "# Ratio of steps for a linear warmup (from 0 to learning rate)\n",
        "warmup_ratio = 0.03\n",
        "\n",
        "# Group sequences into batches with same length\n",
        "# Saves memory and speeds up training considerably\n",
        "group_by_length = True\n",
        "\n",
        "# Save checkpoint every X updates steps\n",
        "save_steps = 0\n",
        "\n",
        "# Log every X updates steps\n",
        "logging_steps = 25\n",
        "\n",
        "################################################################################\n",
        "# SFT parameters\n",
        "################################################################################\n",
        "\n",
        "# Maximum sequence length to use\n",
        "max_seq_length = None\n",
        "\n",
        "# Pack multiple short examples in the same input sequence to increase efficiency\n",
        "packing = False\n",
        "\n",
        "# Load the entire model on the GPU 0\n",
        "device_map = {\"\": 0}"
      ],
      "metadata": {
        "id": "ib_We3NLtj2E"
      },
      "execution_count": 3,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Fine-tuning process**"
      ],
      "metadata": {
        "id": "G9w5Txi5wI2B"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "1. Veri setini yükleme. Veri seti eğer uygun formatta değilse veya birden çok veri seti kullanılacaksa uygun şekilde değiştirilmeli.\n",
        "\n",
        "2. Daha sonra, 4 bitlik quantization için bitsandbytes ları yapılandırıyoruz.\n",
        "\n",
        "\n",
        "3. Daha sonra Llama 2 modelini 4 bit ve  ilgili tokenizer ile GPU'ya yüklüyoruz.\n",
        "\n",
        "\n",
        "4. Son olarak QLoRA için konfigürasyonları, düzenli eğitim parametrelerini yüklüyoruz ve her şeyi SFTTrainer'a aktarıyoruz."
      ],
      "metadata": {
        "id": "qNx4Es23wjzC"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Load dataset (you can process it here)\n",
        "dataset = load_dataset(dataset_name, split=\"train\")\n",
        "\n",
        "# Load tokenizer and model with QLoRA configuration\n",
        "compute_dtype = getattr(torch, bnb_4bit_compute_dtype)\n",
        "\n",
        "bnb_config = BitsAndBytesConfig(\n",
        "    load_in_4bit=use_4bit,\n",
        "    bnb_4bit_quant_type=bnb_4bit_quant_type,\n",
        "    bnb_4bit_compute_dtype=compute_dtype,\n",
        "    bnb_4bit_use_double_quant=use_nested_quant,\n",
        ")\n",
        "\n",
        "# Check GPU compatibility with bfloat16\n",
        "if compute_dtype == torch.float16 and use_4bit:\n",
        "    major, _ = torch.cuda.get_device_capability()\n",
        "    if major >= 8:\n",
        "        print(\"=\" * 80)\n",
        "        print(\"Your GPU supports bfloat16: accelerate training with bf16=True\")\n",
        "        print(\"=\" * 80)\n",
        "\n",
        "# Load base model\n",
        "model = AutoModelForCausalLM.from_pretrained(\n",
        "    model_name,\n",
        "    quantization_config=bnb_config,\n",
        "    device_map=device_map\n",
        ")\n",
        "model.config.use_cache = False\n",
        "model.config.pretraining_tp = 1\n",
        "\n",
        "# Load LLaMA tokenizer\n",
        "tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)\n",
        "tokenizer.pad_token = tokenizer.eos_token\n",
        "tokenizer.padding_side = \"right\" # Fix weird overflow issue with fp16 training\n",
        "\n",
        "# Load LoRA configuration\n",
        "peft_config = LoraConfig(\n",
        "    lora_alpha=lora_alpha,\n",
        "    lora_dropout=lora_dropout,\n",
        "    r=lora_r,\n",
        "    bias=\"none\",\n",
        "    task_type=\"CAUSAL_LM\",\n",
        ")\n",
        "\n",
        "# Set training parameters\n",
        "training_arguments = TrainingArguments(\n",
        "    output_dir=output_dir,\n",
        "    num_train_epochs=num_train_epochs,\n",
        "    per_device_train_batch_size=per_device_train_batch_size,\n",
        "    gradient_accumulation_steps=gradient_accumulation_steps,\n",
        "    optim=optim,\n",
        "    save_steps=save_steps,\n",
        "    logging_steps=logging_steps,\n",
        "    learning_rate=learning_rate,\n",
        "    weight_decay=weight_decay,\n",
        "    fp16=fp16,\n",
        "    bf16=bf16,\n",
        "    max_grad_norm=max_grad_norm,\n",
        "    max_steps=max_steps,\n",
        "    warmup_ratio=warmup_ratio,\n",
        "    group_by_length=group_by_length,\n",
        "    lr_scheduler_type=lr_scheduler_type,\n",
        "    report_to=\"tensorboard\"\n",
        ")\n",
        "\n",
        "# Set supervised fine-tuning parameters\n",
        "trainer = SFTTrainer(\n",
        "    model=model,\n",
        "    train_dataset=dataset,\n",
        "    peft_config=peft_config,\n",
        "    dataset_text_field=\"text\",\n",
        "    max_seq_length=max_seq_length,\n",
        "    tokenizer=tokenizer,\n",
        "    args=training_arguments,\n",
        "    packing=packing,\n",
        ")\n",
        "\n",
        "# Train model\n",
        "trainer.train()\n",
        "\n",
        "# Save trained model\n",
        "trainer.model.save_pretrained(new_model)"
      ],
      "metadata": {
        "id": "OJXpOgBFuSrc",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000,
          "referenced_widgets": [
            "ad3aefa473b24f14ab44d5ab3067cc1a",
            "09bb3b4001514eed8638b2094ee6d903",
            "41db7902d4414b978838c78c624c942e",
            "bac866fb454c42b68f1118fdc0b3b305",
            "c3e033aafe614a1c938a076c9783631e",
            "b6d00ffc2a184a0c86ec46a5f45db26b",
            "b9f4081551a8463b95df7c3ee802a45d",
            "b83529889ab8421184cc0368ca23de5b",
            "3b746c608411451eb577ad73543656ac",
            "e7d66667449f4c65bb9c6b7795d9f4dd",
            "6b2ddab8b63d4b69838463af41aa450b"
          ]
        },
        "outputId": "49a090e7-6d27-4109-c696-3ffbfd2b1fed"
      },
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "ad3aefa473b24f14ab44d5ab3067cc1a"
            }
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/peft/utils/other.py:102: FutureWarning: prepare_model_for_int8_training is deprecated and will be removed in a future version. Use prepare_model_for_kbit_training instead.\n",
            "  warnings.warn(\n",
            "/usr/local/lib/python3.10/dist-packages/trl/trainer/sft_trainer.py:159: UserWarning: You didn't pass a `max_seq_length` argument to the SFTTrainer, this will default to 1024\n",
            "  warnings.warn(\n",
            "You're using a LlamaTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
            "/usr/local/lib/python3.10/dist-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<IPython.core.display.HTML object>"
            ],
            "text/html": [
              "\n",
              "    <div>\n",
              "      \n",
              "      <progress value='1313' max='1313' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
              "      [1313/1313 3:27:22, Epoch 1/1]\n",
              "    </div>\n",
              "    <table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              " <tr style=\"text-align: left;\">\n",
              "      <th>Step</th>\n",
              "      <th>Training Loss</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <td>25</td>\n",
              "      <td>2.365600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>50</td>\n",
              "      <td>1.843900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>75</td>\n",
              "      <td>1.420200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>100</td>\n",
              "      <td>1.204200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>125</td>\n",
              "      <td>1.115900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>150</td>\n",
              "      <td>0.998600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>175</td>\n",
              "      <td>0.992500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>200</td>\n",
              "      <td>0.918900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>225</td>\n",
              "      <td>0.962100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>250</td>\n",
              "      <td>0.905400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>275</td>\n",
              "      <td>0.919600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>300</td>\n",
              "      <td>0.813500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>325</td>\n",
              "      <td>0.920400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>350</td>\n",
              "      <td>0.853300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>375</td>\n",
              "      <td>0.808600</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>400</td>\n",
              "      <td>0.781100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>425</td>\n",
              "      <td>0.823100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>450</td>\n",
              "      <td>0.770700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>475</td>\n",
              "      <td>0.831400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>500</td>\n",
              "      <td>0.752900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>525</td>\n",
              "      <td>0.794800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>550</td>\n",
              "      <td>0.774400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>575</td>\n",
              "      <td>0.770500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>600</td>\n",
              "      <td>0.745100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>625</td>\n",
              "      <td>0.790100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>650</td>\n",
              "      <td>0.689400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>675</td>\n",
              "      <td>0.770100</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>700</td>\n",
              "      <td>0.712800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>725</td>\n",
              "      <td>0.711700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>750</td>\n",
              "      <td>0.751700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>775</td>\n",
              "      <td>0.689900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>800</td>\n",
              "      <td>0.737400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>825</td>\n",
              "      <td>0.714300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>850</td>\n",
              "      <td>0.716800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>875</td>\n",
              "      <td>0.736400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>900</td>\n",
              "      <td>0.664900</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>925</td>\n",
              "      <td>0.716200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>950</td>\n",
              "      <td>0.690400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>975</td>\n",
              "      <td>0.716800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1000</td>\n",
              "      <td>0.662400</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1025</td>\n",
              "      <td>0.700200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1050</td>\n",
              "      <td>0.675800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1075</td>\n",
              "      <td>0.729500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1100</td>\n",
              "      <td>0.720000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1125</td>\n",
              "      <td>0.680200</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1150</td>\n",
              "      <td>0.700000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1175</td>\n",
              "      <td>0.653300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1200</td>\n",
              "      <td>0.666300</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1225</td>\n",
              "      <td>0.706700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1250</td>\n",
              "      <td>0.625000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1275</td>\n",
              "      <td>0.670700</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <td>1300</td>\n",
              "      <td>0.656000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table><p>"
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "##**Tensorboard dan görüntüleyelim**"
      ],
      "metadata": {
        "id": "NBtXo_Tgw43o"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "%load_ext tensorboard\n",
        "%tensorboard --logdir results/runs"
      ],
      "metadata": {
        "id": "crj9svNe4hU5"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Modelin denenmesi**"
      ],
      "metadata": {
        "id": "TDqxnjvExGG6"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Ignore warnings\n",
        "logging.set_verbosity(logging.CRITICAL)\n",
        "\n",
        "# Run text generation pipeline with our next model\n",
        "prompt = \" köfte nasıl yapılır?\"\n",
        "pipe = pipeline(task=\"text-generation\", model=model, tokenizer=tokenizer, max_length=500)\n",
        "result = pipe(f\"<s>[INST] {prompt} [/INST]\")\n",
        "print(result[0]['generated_text'])"
      ],
      "metadata": {
        "id": "frlSLPin4IJ4",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ae1cad58-3bb9-4740-9b70-89ce423718dc"
      },
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<s>[INST]  köfte nasıl yapılır? [/INST] Köfte yapımında; orta boy derin bir tencere içine 1 adet yumurta kırın. (Yumurtanın kırımını azar azar sıcak olacaktır). Ardından 1 su bardağı sıvı yağ, 1 çay bardağı kadar tuz ve 1 çay bardağı kadar kırmızı toz biber ekleyin. Tencerenin kapağını kapatın. Tencerenin içine 100 gr kıyma koyun. Kıymayı 1 dakika kadar kavurun.Daha sonra tencere içine 100 gr kıyma koyun. Kıymayı 1 dakika kadar daha kavurun. Ardından tencere içine 100 gr kıyma koyun. Kıymayı 1 dakika kadar daha kavurun. Son olarak tencere içine 100 gr kıyma koyun. Kıymayı 1 dakika kadar daha kavurun. Tencerenin kapağını kapatın. Köfte harcını 15 dakika kadar kavurun.Diğer tarafta köfte harcının üzerine 1 su bardağı kadar un ekleyin. Köfte harcını 15 dakika kadar daha kavurun. Ardından tencerenin kapağını kapatın. Köfte harcını 15 dakika kadar daha dinlenmeye bırakın.Daha sonra köfte harcını çok fazla dinlenmeden önce 10 dakika kadar\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Yeni Llama2 Modeli**"
      ],
      "metadata": {
        "id": "CwbthYYhxs8p"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Modeli depolamak için LoRA dan gelen ağırlıkları ana modelle birleştirmeliyiz."
      ],
      "metadata": {
        "id": "ugs6EbD9xtAs"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# Reload model in FP16 and merge it with LoRA weights\n",
        "base_model = AutoModelForCausalLM.from_pretrained(\n",
        "    model_name,\n",
        "    low_cpu_mem_usage=True,\n",
        "    return_dict=True,\n",
        "    torch_dtype=torch.float16,\n",
        "    device_map=device_map,\n",
        ")\n",
        "model = PeftModel.from_pretrained(base_model, new_model)\n",
        "model = model.merge_and_unload()\n",
        "\n",
        "# Reload tokenizer to save it\n",
        "tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)\n",
        "tokenizer.pad_token = tokenizer.eos_token\n",
        "tokenizer.padding_side = \"right\""
      ],
      "metadata": {
        "id": "QQn30cRtAZ-P",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 49,
          "referenced_widgets": [
            "cff4c007189341db853f12e91ab926b3",
            "8d57720cb7dd41618193249f3c3a164b",
            "cdca0bbac7ac4c1096f46f4bd0f8e2b6",
            "031089c370934c57b9e631c7db9b247c",
            "ef6f7ae8e41643a0bb372a0869ba059d",
            "523c5f99629d4a258f1f53b63b32c029",
            "3138e4baaaf24cdeacbd09888dd56383",
            "e51e970e57774d589110ea29bf2c258f",
            "f76695423cdb419d80e883c2a34ae1c9",
            "6dda331a137943a59f2e86273f9a4317",
            "eb9e401d0cf948509f5b29a804e73b4c"
          ]
        },
        "outputId": "5a3cfaba-047a-4c05-9f5d-91d0cc0ecd67"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "Loading checkpoint shards:   0%|          | 0/2 [00:00<?, ?it/s]"
            ],
            "application/vnd.jupyter.widget-view+json": {
              "version_major": 2,
              "version_minor": 0,
              "model_id": "cff4c007189341db853f12e91ab926b3"
            }
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "#**Modeli Hugging Face Hub'a yüklemek**"
      ],
      "metadata": {
        "id": "2bn7tjdByJ_i"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "import locale\n",
        "locale.getpreferredencoding = lambda: \"UTF-8\""
      ],
      "metadata": {
        "id": "Z8VUygTdqVnJ"
      },
      "execution_count": 9,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!huggingface-cli login\n",
        "\n",
        "model.push_to_hub(\"mertbozkurt/Llama-2-7b-chat-finetune\", check_pr=True)\n",
        "\n",
        "tokenizer.push_to_hub(\"mertbozkurt/Llama-2-7b-chat-finetune\",check_pr=True)\n"
      ],
      "metadata": {
        "id": "x-xPb-_qB0dz"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}