AML_project_model.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": [],
      "gpuType": "T4"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    },
    "accelerator": "GPU"
  },
  "cells": [
    {
      "cell_type": "markdown",
      "source": [
        "# EE782: AML Project\n",
        "# IITB SmartBot"
      ],
      "metadata": {
        "id": "bpOOdKB1ntAH"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "Dhatri Mehta 210070027</br>\n",
        "Devesh Soni 21D070025</br>\n",
        "Shivansh Gupta 21D070067"
      ],
      "metadata": {
        "id": "JKeuFoeLntdf"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Packages and Libraries"
      ],
      "metadata": {
        "id": "3LIlUgXYni4v"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "## This was just to check the status and usage of compute unit of GPU\n",
        "!nvidia-smi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AFZMjbPtyXl8",
        "outputId": "44ee5c07-6cbc-4f44-d8a0-a02f51b0a947"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Thu Nov 30 10:32:59 2023       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 525.105.17   Driver Version: 525.105.17   CUDA Version: 12.0     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|                               |                      |               MIG M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla T4            Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   62C    P8    11W /  70W |      0MiB / 15360MiB |      0%      Default |\n",
            "|                               |                      |                  N/A |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                                  |\n",
            "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
            "|        ID   ID                                                   Usage      |\n",
            "|=============================================================================|\n",
            "|  No running processes found                                                 |\n",
            "+-----------------------------------------------------------------------------+\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "eQlQ4ZWgcXDW",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2305bec2-382f-4df4-c601-85d70130fe01"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m\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.\n",
            "torchaudio 2.1.0+cu118 requires torch==2.1.0, but you have torch 2.0.1 which is incompatible.\n",
            "torchdata 0.7.0 requires torch==2.1.0, but you have torch 2.0.1 which is incompatible.\n",
            "torchtext 0.16.0 requires torch==2.1.0, but you have torch 2.0.1 which is incompatible.\n",
            "torchvision 0.16.0+cu118 requires torch==2.1.0, but you have torch 2.0.1 which is incompatible.\u001b[0m\u001b[31m\n",
            "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m\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.\n",
            "xformers 0.0.20 requires torch==2.0.1, but you have torch 2.1.0 which is incompatible.\u001b[0m\u001b[31m\n",
            "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m"
          ]
        }
      ],
      "source": [
        "# This code installs specific versions of various Python packages using pip, suppressing progress bars for a cleaner output\n",
        "\n",
        "!pip install -Uqqq pip --progress-bar off\n",
        "!pip install -qqq torch==2.0.1 --progress-bar off\n",
        "!pip install -qqq transformers==4.31.0 --progress-bar off\n",
        "!pip install -qqq langchain==0.0.266 --progress-bar off\n",
        "!pip install -qqq chromadb==0.4.5 --progress-bar off\n",
        "!pip install -qqq pypdf==3.15.0 --progress-bar off\n",
        "!pip install -qqq xformers==0.0.20 --progress-bar off\n",
        "!pip install -qqq sentence_transformers==2.2.2 --progress-bar off\n",
        "!pip install -qqq InstructorEmbedding==1.0.1 --progress-bar off\n",
        "!pip install -qqq pdf2image==1.16.3 --progress-bar off"
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Binary wheel files in Python are pre-built distributions of a package that can be easily installed on a compatible system without the need for compilation. They play a crucial role in simplifying the installation process and improving the efficiency of package distribution. For our model we are downloading a AUTOGPTQ BWF file package"
      ],
      "metadata": {
        "id": "QPBtTlTem3LK"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "#using the wget command to download a binary wheel file\n",
        "#The file being downloaded is \"auto_gptq-0.4.1+cu118-cp310-cp310-linux_x86_64.whl,\"  related to a AutoGPTQ model\n",
        "\n",
        "!wget -q https://github.com/PanQiWei/AutoGPTQ/releases/download/v0.4.1/auto_gptq-0.4.1+cu118-cp310-cp310-linux_x86_64.whl"
      ],
      "metadata": {
        "id": "ecTRR8hwhn0C"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "!pip install -qqq auto_gptq-0.4.1+cu118-cp310-cp310-linux_x86_64.whl --progress-bar off"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "C0_Fa50LkfoB",
        "outputId": "ec2881e5-7fcd-43a0-843a-be58450e879a"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
            "\u001b[0m"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!apt-get update"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BpfpTk4SO-XU",
        "outputId": "65f31919-5c0d-406b-f07a-daa0973ed052"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\r0% [Working]\r            \rHit:1 https://cloud.r-project.org/bin/linux/ubuntu jammy-cran40/ InRelease\n",
            "Hit:2 https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64  InRelease\n",
            "Hit:3 http://archive.ubuntu.com/ubuntu jammy InRelease\n",
            "Hit:4 http://archive.ubuntu.com/ubuntu jammy-updates InRelease\n",
            "Hit:5 http://archive.ubuntu.com/ubuntu jammy-backports InRelease\n",
            "Hit:6 http://security.ubuntu.com/ubuntu jammy-security InRelease\n",
            "Hit:7 https://ppa.launchpadcontent.net/c2d4u.team/c2d4u4.0+/ubuntu jammy InRelease\n",
            "Hit:8 https://ppa.launchpadcontent.net/deadsnakes/ppa/ubuntu jammy InRelease\n",
            "Hit:9 https://ppa.launchpadcontent.net/graphics-drivers/ppa/ubuntu jammy InRelease\n",
            "Hit:10 https://ppa.launchpadcontent.net/ubuntugis/ppa/ubuntu jammy InRelease\n",
            "Reading package lists... Done\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!apt-get install poppler-utils"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7eUqwK5HPDox",
        "outputId": "4844a010-aab3-4b4d-fd6b-b0d1c0f4bc8f"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Reading package lists... Done\n",
            "Building dependency tree... Done\n",
            "Reading state information... Done\n",
            "poppler-utils is already the newest version (22.02.0-2ubuntu0.3).\n",
            "0 upgraded, 0 newly installed, 0 to remove and 19 not upgraded.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!sudo apt-get install poppler-utils"
      ],
      "metadata": {
        "id": "t7ShwfeTubDU",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ec60c9f4-0320-4b51-e339-4589b6746693"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Reading package lists... Done\n",
            "Building dependency tree... Done\n",
            "Reading state information... Done\n",
            "poppler-utils is already the newest version (22.02.0-2ubuntu0.3).\n",
            "0 upgraded, 0 newly installed, 0 to remove and 19 not upgraded.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## importing various libraries required for our LLM\n",
        "import torch\n",
        "from auto_gptq import AutoGPTQForCausalLM\n",
        "from langchain import HuggingFacePipeline, PromptTemplate\n",
        "from langchain.chains import RetrievalQA\n",
        "from langchain.document_loaders import PyPDFDirectoryLoader\n",
        "from langchain.embeddings import HuggingFaceInstructEmbeddings\n",
        "from langchain.text_splitter import RecursiveCharacterTextSplitter\n",
        "from langchain.vectorstores import Chroma\n",
        "from pdf2image import convert_from_path\n",
        "from transformers import AutoTokenizer, TextStreamer, pipeline\n",
        "\n",
        "DEVICE = \"cuda:0\" if torch.cuda.is_available() else \"cpu\""
      ],
      "metadata": {
        "id": "vH5IpjeNflMU"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Data"
      ],
      "metadata": {
        "id": "jiX57tzWOJAg"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "### The below code show how our pdf looks like"
      ],
      "metadata": {
        "id": "JwObEE0GJELD"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "scraped_data_images = convert_from_path(\"/content/Scraped_data/Final_data.pdf\", dpi=88)\n",
        "scraped_data_images[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 986
        },
        "id": "a9X4Fbv8ugOW",
        "outputId": "c792e917-e4c4-4906-8097-740d9c6680b0"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<PIL.PpmImagePlugin.PpmImageFile image mode=RGB size=749x969>"
            ],
            "image/png": "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\n"
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!rm -rf \"db\""
      ],
      "metadata": {
        "id": "0A_9bQ7-N3d6"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "## loading the pdf file\n",
        "loader = PyPDFDirectoryLoader(\"/content/Scraped_data\")\n",
        "docs = loader.load()\n",
        "len(docs)                        # number of pages in the pdf"
      ],
      "metadata": {
        "id": "RQmhREpeOInh",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "860ce5df-1c0b-49e6-b424-aed76b2c1bb3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "82"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "##  this code is setting up an embedding generator using the Hugging Face \"instructor-large\" model\n",
        "embeddings = HuggingFaceInstructEmbeddings(\n",
        "    model_name=\"hkunlp/instructor-large\", model_kwargs={\"device\": DEVICE}\n",
        ")"
      ],
      "metadata": {
        "id": "wjJ0DI9MRNkT",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c8e551a3-3778-46b4-d32b-bf6b4022ef8b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "load INSTRUCTOR_Transformer\n",
            "max_seq_length  512\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "This code is using a recursive character-based text splitter to divide a collection of documents into smaller chunks of text\n",
        "One chunk contains 1024 words and each chunk has an overlap of 64 words with successive and predecessor chunk. This help to form a proper connection between differnet chunks helping the model to learn efficiently."
      ],
      "metadata": {
        "id": "xnWxaUR0rXfw"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "# this code is using a recursive character-based text splitter to divide a collection of documents into smaller chunks of text\n",
        "\n",
        "text_splitter = RecursiveCharacterTextSplitter(chunk_size=1024, chunk_overlap=64)\n",
        "texts = text_splitter.split_documents(docs)\n",
        "len(texts)"
      ],
      "metadata": {
        "id": "yFztsob3WO3x",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9216d5df-a9cc-4f62-ef10-2a60344991e1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "164"
            ]
          },
          "metadata": {},
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Chroma** is an open-source embedding database that is used to store and manage embeddings. </br>\n",
        "This code creates a Chroma object by calling the from_documents method. The method takes texts (a list of text chunks created above), embeddings (hkunlp/HuggingFace), and persist_directory=\"db\""
      ],
      "metadata": {
        "id": "Hd15a3c7sOeI"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "db = Chroma.from_documents(texts, embeddings, persist_directory=\"db\")"
      ],
      "metadata": {
        "id": "ESFJOOYEWTX9"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Llama 2 13B"
      ],
      "metadata": {
        "id": "GcylvlF60r_m"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "This code snippet is initializing a language model using the Hugging Face Transformers library</br>\n",
        "The pre-trained language model we are using is **TheBloke/Llama-2**-13B-chat-GPTQ.</br>\n",
        "The model variable now holds an instance of the **GPTQ language model** ready for use for our Q&A tasks."
      ],
      "metadata": {
        "id": "5AHDHhAltgPS"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "model_name_or_path = \"TheBloke/Llama-2-13B-chat-GPTQ\"\n",
        "model_basename = \"model\"\n",
        "\n",
        "#  initializes a tokenizer using the AutoTokenizer\n",
        "tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)\n",
        "\n",
        "# This initializes the GPTQ (Quantized GPT) language model using the AutoGPTQForCausalLM class\n",
        "# The from_quantized method is used to load a quantized version of the model, which we have downloaded above as a binary wheel file\n",
        "model = AutoGPTQForCausalLM.from_quantized(\n",
        "    model_name_or_path,\n",
        "    revision=\"gptq-4bit-128g-actorder_True\",\n",
        "    model_basename=model_basename,\n",
        "    use_safetensors=True,\n",
        "    trust_remote_code=True,\n",
        "    inject_fused_attention=False,\n",
        "    device=DEVICE,\n",
        "    quantize_config=None,\n",
        ")"
      ],
      "metadata": {
        "id": "DrSTLMQbfjFp",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "68792769-0468-4b37-e056-b8136b506505"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:auto_gptq.nn_modules.fused_llama_mlp:skip module injection for FusedLlamaMLPForQuantizedModel not support integrate without triton yet.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "## This was just to check the status and usage of compute unit of GPU\n",
        "!nvidia-smi"
      ],
      "metadata": {
        "id": "k6yLtagJfvG4",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "6763e7fd-f09f-4b5f-b659-740bc811cd19"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Thu Nov 30 10:37:10 2023       \n",
            "+-----------------------------------------------------------------------------+\n",
            "| NVIDIA-SMI 525.105.17   Driver Version: 525.105.17   CUDA Version: 12.0     |\n",
            "|-------------------------------+----------------------+----------------------+\n",
            "| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |\n",
            "| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |\n",
            "|                               |                      |               MIG M. |\n",
            "|===============================+======================+======================|\n",
            "|   0  Tesla T4            Off  | 00000000:00:04.0 Off |                    0 |\n",
            "| N/A   58C    P0    37W /  70W |   9593MiB / 15360MiB |     60%      Default |\n",
            "|                               |                      |                  N/A |\n",
            "+-------------------------------+----------------------+----------------------+\n",
            "                                                                               \n",
            "+-----------------------------------------------------------------------------+\n",
            "| Processes:                                                                  |\n",
            "|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |\n",
            "|        ID   ID                                                   Usage      |\n",
            "|=============================================================================|\n",
            "+-----------------------------------------------------------------------------+\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "This is **System Prompt** structure which we have designed for use in a context to guide the behavior of our language model with specific instructions while taking into account user prompts."
      ],
      "metadata": {
        "id": "BjRk9jG_0On1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "DEFAULT_SYSTEM_PROMPT = \"\"\"\n",
        "You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature.\n",
        "\n",
        "If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.\n",
        "\"\"\".strip()\n",
        "\n",
        "def generate_prompt(prompt: str, system_prompt: str = DEFAULT_SYSTEM_PROMPT) -> str:\n",
        "    return f\"\"\"\n",
        "[INST] <<SYS>>\n",
        "{system_prompt}\n",
        "<</SYS>>\n",
        "\n",
        "{prompt} [/INST]\n",
        "\"\"\".strip()"
      ],
      "metadata": {
        "id": "NKUo_0gxRD-a"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "The **TextStreamer** class is a utility class from the transformers library that is used to stream the output of a language model during generation. This can be useful for applications where you want to display the generated text as it is being produced, or for collecting the output of the model without waiting for the entire generation to finish."
      ],
      "metadata": {
        "id": "5VaSrEj01Rx1"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)"
      ],
      "metadata": {
        "id": "5H6rPBpEAALY"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "## This code is creating a text generation pipeline\n",
        "text_pipeline = pipeline(\n",
        "    \"text-generation\",\n",
        "    model=model,                                                        # pre-trained language model to be used for text generation\n",
        "    tokenizer=tokenizer,                                                # tokenizer defined above\n",
        "    max_new_tokens=1024,                                                # Sets the maximum number of tokens that can be generated as output\n",
        "    temperature=0,                                                      # Temperature is a hyperparameter that controls the randomness of the generated text.\n",
        "                                                                           # A temperature of 0 results in deterministic output, meaning the most likely token is always chosen.\n",
        "    top_p=0.95,                                                         # controls the diversity of the generated output\n",
        "    repetition_penalty=1.15,                                            # Repetition penalty discourages the model from repeating the same tokens in its output\n",
        "    streamer=streamer,\n",
        ")\n"
      ],
      "metadata": {
        "id": "ZYaQDkDvVvIz",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a2894e56-881e-4f71-9bf2-b459f1c953f6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "WARNING:xformers:WARNING[XFORMERS]: xFormers can't load C++/CUDA extensions. xFormers was built for:\n",
            "    PyTorch 2.0.1+cu118 with CUDA 1108 (you have 2.1.0+cu121)\n",
            "    Python  3.10.11 (you have 3.10.12)\n",
            "  Please reinstall xformers (see https://github.com/facebookresearch/xformers#installing-xformers)\n",
            "  Memory-efficient attention, SwiGLU, sparse and more won't be available.\n",
            "  Set XFORMERS_MORE_DETAILS=1 for more details\n",
            "The model 'LlamaGPTQForCausalLM' is not supported for text-generation. Supported models are ['BartForCausalLM', 'BertLMHeadModel', 'BertGenerationDecoder', 'BigBirdForCausalLM', 'BigBirdPegasusForCausalLM', 'BioGptForCausalLM', 'BlenderbotForCausalLM', 'BlenderbotSmallForCausalLM', 'BloomForCausalLM', 'CamembertForCausalLM', 'CodeGenForCausalLM', 'CpmAntForCausalLM', 'CTRLLMHeadModel', 'Data2VecTextForCausalLM', 'ElectraForCausalLM', 'ErnieForCausalLM', 'FalconForCausalLM', 'GitForCausalLM', 'GPT2LMHeadModel', 'GPT2LMHeadModel', 'GPTBigCodeForCausalLM', 'GPTNeoForCausalLM', 'GPTNeoXForCausalLM', 'GPTNeoXJapaneseForCausalLM', 'GPTJForCausalLM', 'LlamaForCausalLM', 'MarianForCausalLM', 'MBartForCausalLM', 'MegaForCausalLM', 'MegatronBertForCausalLM', 'MusicgenForCausalLM', 'MvpForCausalLM', 'OpenLlamaForCausalLM', 'OpenAIGPTLMHeadModel', 'OPTForCausalLM', 'PegasusForCausalLM', 'PLBartForCausalLM', 'ProphetNetForCausalLM', 'QDQBertLMHeadModel', 'ReformerModelWithLMHead', 'RemBertForCausalLM', 'RobertaForCausalLM', 'RobertaPreLayerNormForCausalLM', 'RoCBertForCausalLM', 'RoFormerForCausalLM', 'RwkvForCausalLM', 'Speech2Text2ForCausalLM', 'TransfoXLLMHeadModel', 'TrOCRForCausalLM', 'XGLMForCausalLM', 'XLMWithLMHeadModel', 'XLMProphetNetForCausalLM', 'XLMRobertaForCausalLM', 'XLMRobertaXLForCausalLM', 'XLNetLMHeadModel', 'XmodForCausalLM'].\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# creating an instance of the HuggingFacePipeline class, for a pipeline for text generation (text_pipeline) created above\n",
        "llm = HuggingFacePipeline(pipeline=text_pipeline, model_kwargs={\"temperature\": 0})"
      ],
      "metadata": {
        "id": "s0sKzP77ZCok"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "SYSTEM_PROMPT = \"Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer.\"\n",
        "\n",
        "template = generate_prompt(\n",
        "    \"\"\"\n",
        "{context}\n",
        "\n",
        "Question: {question}\n",
        "\"\"\",\n",
        "    system_prompt=SYSTEM_PROMPT,\n",
        ")"
      ],
      "metadata": {
        "id": "J3RKlMe4DNMn"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "prompt = PromptTemplate(template=template, input_variables=[\"context\", \"question\"])"
      ],
      "metadata": {
        "id": "pJpTuSEoDDEJ"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# This code is creating a RetrievalQA object named qa_chain using the question-answering pipeline (llm)\n",
        "qa_chain = RetrievalQA.from_chain_type(\n",
        "    llm=llm,\n",
        "    chain_type=\"stuff\",\n",
        "    retriever=db.as_retriever(search_kwargs={\"k\": 2}),\n",
        "    return_source_documents=True,\n",
        "    chain_type_kwargs={\"prompt\": prompt},\n",
        ")"
      ],
      "metadata": {
        "id": "0DC-RU6DZCIS"
      },
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Chat with our SmartBOT"
      ],
      "metadata": {
        "id": "J7IWZ6H1ax4z"
      }
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"hello how are you\")"
      ],
      "metadata": {
        "id": "1nvSY7aZvDKq",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "2dbf621b-e6ef-4862-c72e-23a7b5a200ce"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Hello! I'm doing well, thank you for asking. However, I would like to point out that the question is a bit ambiguous and open-ended. Could you please provide more context or clarify what you mean by \"hello how are you\"? I want to make sure I give you an accurate and helpful response.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"I have some questions, can I ask you?\")"
      ],
      "metadata": {
        "id": "KrFtfkgPvH6Q",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "8ac92031-319b-4133-e5e6-542dd6913b24"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Sure! I'll do my best to help answer your questions. What are your questions?\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"What is SMP?\")"
      ],
      "metadata": {
        "id": "nKmLM4UTfVmL",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "c94a6747-8376-4555-ddc5-8463bad366d1"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Based on the given context, I can answer the following question:\n",
            "\n",
            "What is SMP?\n",
            "\n",
            "SMP stands for Student Mentor Programme, which is a program within IIT Bombay that aims to provide constructive and positive interaction, guidance, and mentorship to junior students by senior students. The primary objective of SMP is to motivate students to excel in both academic and non-academic fields and make the most of their life at IIT Bombay.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"How can I book a room in guest house?\")"
      ],
      "metadata": {
        "id": "1k96dxrgfjMr",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0a1c2b48-f61d-40e5-b1ef-07ecd2993420"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Based on the provided information, here is the answer to your question:\n",
            "\n",
            "To book a room in the Guest House, you can obtain a booking form from the Hostel Coordinating Unit (HCU) located in the Main Building (MB). You may also email hcu.office@iitb.ac.in to request a form. Once you have filled out the form and obtained approval, take the approved form to the Hall Manager of the allotted hostel and collect the key from their office. In emergency situations, guests can stay in the hostel rooms after entry in the Guest Register kept with the security at the hostel entrance. The formalities can be completed on the next working day.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"What are the different scholarships provided at iit Bomaby?\")"
      ],
      "metadata": {
        "id": "waIck-yefydp",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a7e2166e-2e0e-442a-9455-1e33cd281c15"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Based on the information provided, there are two types of scholarships available at IIT Bombay:\n",
            "\n",
            "1. Merit cum Means Scholarship: This scholarship is based on both merit (JEE AIR and UCEED rank for B.Tech and B.Des students respectively) and means (parental income). Students whose parents' annual gross income does not exceed Rs. 5,00,000/ are eligible to apply.\n",
            "2. IT Support Scholarship: This scholarship is a financial support scheme for economically weak students, which covers tuition fees, mess advance, and other expenses. It is offered by the Director of IIT Bombay in collaboration with generous alumni of IIT Bombay and the Dean of the ACR office. The scholarship is awarded based on academic performance, and if a student's performance is poor, the Dean may cancel the scholarship.\n",
            "\n",
            "There is also an AP Financial Aid Programme* with a 0% interest rate, offered by Tata Motors for UGs and PGs whose annual family income falls below 6.5 lakh per annum, provided the student has a CPI > 6. The student can contact the Alumni office through an online portal to apply. Documents required for application include 12th mark sheets, JEE advanced results, family income certificate, parents residential proof, and parents ID proof. The interview with the FAP committee (IIT-B Alumni) follows after the application process. Various kinds of assistance provided under this programme include tuition and mess fee, purchasing a laptop, and internship expenses. The repayment is done after placement, with a grace period of 6 months.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"How can I calculate CPI and SPI?\")"
      ],
      "metadata": {
        "id": "nq6cf8vof6wM",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "eccaf885-603c-4ba7-96bc-d57f5c70143e"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Based on the given context, I can answer the following questions:\n",
            "\n",
            "1. How to calculate SPI (Semester Performance Index):\n",
            "To calculate SPI, we need to multiply the credit points earned in each course by the grade points obtained in that course, and then divide the total credit points earned by the total credits taken.\n",
            "\n",
            "Formula: SPI = (C1 x G1 + C2 x G2 + C3 x G3 + C4 x G4) / (C1 + C2 + C3 + C4)\n",
            "\n",
            "Where:\n",
            "\n",
            "* C1, C2, C3, and C4 are the credits earned in each course.\n",
            "* G1, G2, G3, and G4 are the grade points obtained in each course.\n",
            "2. How to calculate CPI (Cumulative Performance Index):\n",
            "To calculate CPI, we need to multiply the SPI for each semester by the number of credits taken in that semester, and then divide the total CPI by the total credits taken.\n",
            "\n",
            "Formula: CPI = (SPI1 x Credits in Semester 1 + SPI2 x Credits in Semester 2 + ...) / (Total Credits)\n",
            "\n",
            "Where:\n",
            "\n",
            "* SPI1 and SPI2 are the SPI values for each semester.\n",
            "* Credits in Semester 1 and Credits in Semester 2 are the number of credits taken in each semester.\n",
            "\n",
            "Note: All semesters have equal credit weightage, so it is important to do well throughout the academic year, not just in final exams, to maintain a good CPI.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"what is Mood Indigo and what all should I need to know to attend it?\")"
      ],
      "metadata": {
        "id": "SIFTDoxzgY07",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "9f23ff28-f3af-4a66-9b3a-59141239afc6"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Based on the given context, here are the answers to your questions:\n",
            "\n",
            "1. What is Mood Indigo?\n",
            "Mood Indigo is a college festival that offers exposure, mentorship, network, funding opportunities, and wisdom to turn dreams into reality for college students.\n",
            "2. What all should I need to know to attend Mood Indigo?\n",
            "To attend Mood Indigo, you will need to have an MI number, which you can obtain by registering on the website. You will also need a college ID to participate in certain activities or competitions. The entry to the festival is free for all college students. The artist lineup for Mood Indigo 2022 has not been announced yet, but you can stay tuned to their official page for updates. After completing the registration, you should note down your unique MI number, as it will be used to identify yourself while attending the festival.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"When is Mood Indigo?\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LLa55clXvpFG",
        "outputId": "419c8c6f-7afd-41e2-c9b8-5a0ede565c68"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Based on the information provided, Mood Indigo will be held from December 18th to December 21st, 2023.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\n",
        "    \"what is Semester Excahnge?\"\n",
        ")"
      ],
      "metadata": {
        "id": "cxp167huggFb",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "3876a1bf-1eb5-4d0d-c148-991735f327ec"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Based on the provided text, I can answer the following question:\n",
            "\n",
            "What is Semester Exchange?\n",
            "\n",
            "According to the text, Semester Exchange is a program offered by IIT Bombay that allows students to study abroad at partner universities across the globe while still being a student at IIT Bombay. Students can spend up to two semesters abroad, taking courses that they would otherwise have to finish at IIT Bombay. The institute has partnerships with numerous universities, and students do not need to pay tuition fees at the foreign university. Additionally, students can explore a wide range of courses that would otherwise be inaccessible, and engaging in research projects expands their knowledge base.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"Who is GSHA and can you provide his contact?\")"
      ],
      "metadata": {
        "id": "vRuJYwSbgpij",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "ad361f00-a44e-4972-dfe7-d4f64fcc090b"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stderr",
          "text": [
            "/usr/local/lib/python3.10/dist-packages/transformers/pipelines/base.py:1083: UserWarning: You seem to be using the pipelines sequentially on GPU. In order to maximize efficiency please use a dataset\n",
            "  warnings.warn(\n"
          ]
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Based on the provided context, GSHA stands for General Secretary of Hostel Affairs at Indian Institute of Technology Bombay (IITB). The name of the person holding this position is Pratham Kapure. His contact information is gsecha@iitb.ac.in.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "result = qa_chain(\"I'm hungry, Where can I get a pizza?\")"
      ],
      "metadata": {
        "id": "X8fsfAq_WQhm",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "617da67d-e511-43f9-d049-e80e3dffafc5"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            " Based on the information provided, there are several options for getting a pizza near IIT Bombay. Here are a few suggestions:\n",
            "\n",
            "1. Dominos: Located in SAC Parking lot, Dominos offers exotic and spicy pizzas with free delivery on campus. You can expect a cheese burst pizza in no more than 30 minutes. Phone: 1800 208 1234. Menu link: m.dominos.co.in\n",
            "2. KFC: Located in the Galleria food court, KFC is one of the many options available for eating out when on a budget. They offer a variety of menu items, including pizzas.\n",
            "3. Pizza Hut: Also located in the Galleria food court, Pizza Hut is another option for pizza lovers. They offer a range of pizza varieties and other menu items.\n",
            "4. Papa Johns: Located in the same area as KFC and Pizza Hut, Papa Johns is another popular choice for pizza. They offer a variety of pizza options and free delivery on orders above a certain amount.\n",
            "\n",
            "Please note that these options are based on the information provided and may not be comprehensive or up-to-date. It's always a good idea to check for current availability and pricing before heading out to eat.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## References\n"
      ],
      "metadata": {
        "id": "hdyKOEdRg2hc"
      }
    },
    {
      "cell_type": "markdown",
      "source": [
        "https://huggingface.co/TheBloke/Llama-2-13B-chat-GPTQ </br>\n",
        "https://huggingface.co/hkunlp/instructor-large </br>\n",
        "https://www.mlexpert.io/prompt-engineering/private-gpt4all#create-chain\n",
        "\n"
      ],
      "metadata": {
        "id": "-UDCfeymoGbj"
      }
    }
  ]
}