Fine_Tuning_LLM.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "source": [
        "!pip install gradientai --upgrade"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "98xwVCRnQjud",
        "outputId": "2232de91-9883-4623-bf72-88f4023662d1"
      },
      "execution_count": 1,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Collecting gradientai\n",
            "  Downloading gradientai-1.13.0-py3-none-any.whl (397 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m397.7/397.7 kB\u001b[0m \u001b[31m4.0 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hCollecting aenum>=3.1.11 (from gradientai)\n",
            "  Downloading aenum-3.1.15-py3-none-any.whl (137 kB)\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m137.6/137.6 kB\u001b[0m \u001b[31m9.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25hRequirement already satisfied: pydantic<3,>=1.10.15 in /usr/local/lib/python3.10/dist-packages (from gradientai) (2.7.3)\n",
            "Requirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.10/dist-packages (from gradientai) (2.8.2)\n",
            "Requirement already satisfied: urllib3>=1.25.3 in /usr/local/lib/python3.10/dist-packages (from gradientai) (2.0.7)\n",
            "Requirement already satisfied: annotated-types>=0.4.0 in /usr/local/lib/python3.10/dist-packages (from pydantic<3,>=1.10.15->gradientai) (0.7.0)\n",
            "Requirement already satisfied: pydantic-core==2.18.4 in /usr/local/lib/python3.10/dist-packages (from pydantic<3,>=1.10.15->gradientai) (2.18.4)\n",
            "Requirement already satisfied: typing-extensions>=4.6.1 in /usr/local/lib/python3.10/dist-packages (from pydantic<3,>=1.10.15->gradientai) (4.12.2)\n",
            "Requirement already satisfied: six>=1.5 in /usr/local/lib/python3.10/dist-packages (from python-dateutil>=2.8.2->gradientai) (1.16.0)\n",
            "Installing collected packages: aenum, gradientai\n",
            "Successfully installed aenum-3.1.15 gradientai-1.13.0\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "import os\n",
        "os.environ['GRADIENT_WORKSPACE_ID'] = 'e73f40fd-a526-455d-9eb9-51da5f8e9633_workspace'\n",
        "os.environ['GRADIENT_ACCESS_TOKEN'] = 'DDjTwb3sK6kIJ07pW3wnmCdQVkZ3aNDm'"
      ],
      "metadata": {
        "id": "OPY7VQwZQ9sR"
      },
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "from gradientai import Gradient\n",
        "\n",
        "\n",
        "def main():\n",
        "    gradient = Gradient()\n",
        "\n",
        "    base_model = gradient.get_base_model(base_model_slug=\"nous-hermes2\")\n",
        "\n",
        "    new_model_adapter = base_model.create_model_adapter(\n",
        "        name=\"myModel\"\n",
        "    )\n",
        "    print(f\"Created model adapter with id {new_model_adapter.id}\")\n",
        "\n",
        "\n",
        "    sample_query = \"### Instruction: who is Chandima Maduwantha? \\n\\n ### Response:\"\n",
        "    print(f\"Asking: {sample_query}\")\n",
        "\n",
        "\n",
        "    ## Before Finetuning\n",
        "    completion = new_model_adapter.complete(query=sample_query, max_generated_token_count=100).generated_output\n",
        "    print(f\"Generated(Before fine tuning): {completion}\")\n",
        "\n",
        "    samples = [\n",
        "        {\"inputs\":\"### Instruction: Who is Chandima Maduwantha? \\n\\n### Response:Chandima Maduwantha is a boy was born in 2000\"},\n",
        "        {\"inputs\":\"### Instruction: Who is this person named Chandima Maduwantha? \\n\\n### Response: Chandima Maduwantha is a undergaduate at the University of Kelaniya in Computer Science.\"},\n",
        "        {\"inputs\":\"### Instruction: What do you know about Chandima Maduwantha? \\n\\n### Response: Chandima is a Software Engineeer who loves Machine Learning and Data Science \"},\n",
        "        ]\n",
        "\n",
        "    ## Lets define parameters for finetuning\n",
        "    num_epochs=3\n",
        "    count=0\n",
        "    while count<num_epochs:\n",
        "      print(f\"Fine tuning the model with iteration {count + 1}\")\n",
        "      new_model_adapter.fine_tune(samples=samples)\n",
        "      count=count+1\n",
        "\n",
        "\n",
        "    #after fine tuning\n",
        "    completion = new_model_adapter.complete(query=sample_query, max_generated_token_count=100).generated_output\n",
        "    print(f\"Generated(after fine tuning): {completion}\")\n",
        "    new_model_adapter.delete()\n",
        "    gradient.close()\n",
        "\n",
        "    # new_model_adapter.delete()\n",
        "    gradient.close()\n",
        "\n",
        "\n",
        "if __name__ == \"__main__\":\n",
        "    main()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cBZz-eIERnhf",
        "outputId": "33cbd850-c4fc-42ff-cff7-542808aee566"
      },
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Created model adapter with id 04dd78bb-9cb7-4def-bbe7-fded6097c4b7_model_adapter\n",
            "Asking: ### Instruction: who is Chandima Maduwantha? \n",
            "\n",
            " ### Response:\n",
            "Generated(Before fine tuning):  Chandima Maduwantha is a Sri Lankan cricketer who plays for the Sri Lanka national cricket team. He is a right-handed batsman and a right-arm off-break bowler. Maduwantha made his first-class debut for Sri Lanka Air Force Sports Club in the 2016–17 Premier League Tournament on 17 March 2017. He has also played for the Sri Lanka\n",
            "Fine tuning the model with iteration 1\n",
            "Fine tuning the model with iteration 2\n",
            "Fine tuning the model with iteration 3\n",
            "Generated(after fine tuning):  Chandima Maduwantha is a boy who was born in 2000 and is currently 21 years old. He is a resident of Sri Lanka and is a student at the University of Kelaniya in Computer Science.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "E3uVAImNMRgH"
      },
      "execution_count": null,
      "outputs": []
    }
  ]
}