LLaVA_13b_4bit_vanilla_colab.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "view-in-github"
      },
      "source": [
        "[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/camenduru/LLaVA-colab/blob/main/LLaVA_13b_4bit_vanilla_colab.ipynb)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "%cd /content\n",
        "!git clone -b v1.0 https://github.com/camenduru/LLaVA\n",
        "%cd /content/LLaVA\n",
        "\n",
        "!pip install -q transformers==4.36.2\n",
        "!pip install -q gradio .\n",
        "\n",
        "from transformers import AutoTokenizer, BitsAndBytesConfig\n",
        "from llava.model import LlavaLlamaForCausalLM\n",
        "import torch\n",
        "\n",
        "model_path = \"4bit/llava-v1.5-13b-3GB\"\n",
        "kwargs = {\"device_map\": \"auto\"}\n",
        "kwargs['load_in_4bit'] = True\n",
        "kwargs['quantization_config'] = BitsAndBytesConfig(\n",
        "    load_in_4bit=True,\n",
        "    bnb_4bit_compute_dtype=torch.float16,\n",
        "    bnb_4bit_use_double_quant=True,\n",
        "    bnb_4bit_quant_type='nf4'\n",
        ")\n",
        "model = LlavaLlamaForCausalLM.from_pretrained(model_path, low_cpu_mem_usage=True, **kwargs)\n",
        "tokenizer = AutoTokenizer.from_pretrained(model_path, use_fast=False)\n",
        "\n",
        "vision_tower = model.get_vision_tower()\n",
        "if not vision_tower.is_loaded:\n",
        "    vision_tower.load_model()\n",
        "vision_tower.to(device='cuda')\n",
        "image_processor = vision_tower.image_processor\n",
        "\n",
        "import os\n",
        "import requests\n",
        "from PIL import Image\n",
        "from io import BytesIO\n",
        "from llava.conversation import conv_templates, SeparatorStyle\n",
        "from llava.utils import disable_torch_init\n",
        "from llava.constants import IMAGE_TOKEN_INDEX, DEFAULT_IMAGE_TOKEN, DEFAULT_IM_START_TOKEN, DEFAULT_IM_END_TOKEN\n",
        "from llava.mm_utils import tokenizer_image_token, get_model_name_from_path, KeywordsStoppingCriteria\n",
        "from transformers import TextStreamer\n",
        "\n",
        "def caption_image(image_file, prompt):\n",
        "    if image_file.startswith('http') or image_file.startswith('https'):\n",
        "        response = requests.get(image_file)\n",
        "        image = Image.open(BytesIO(response.content)).convert('RGB')\n",
        "    else:\n",
        "        image = Image.open(image_file).convert('RGB')\n",
        "    disable_torch_init()\n",
        "    conv_mode = \"llava_v0\"\n",
        "    conv = conv_templates[conv_mode].copy()\n",
        "    roles = conv.roles\n",
        "    image_tensor = image_processor.preprocess(image, return_tensors='pt')['pixel_values'].half().cuda()\n",
        "    inp = f\"{roles[0]}: {prompt}\"\n",
        "    inp = DEFAULT_IM_START_TOKEN + DEFAULT_IMAGE_TOKEN + DEFAULT_IM_END_TOKEN + '\\n' + inp\n",
        "    conv.append_message(conv.roles[0], inp)\n",
        "    conv.append_message(conv.roles[1], None)\n",
        "    raw_prompt = conv.get_prompt()\n",
        "    input_ids = tokenizer_image_token(raw_prompt, tokenizer, IMAGE_TOKEN_INDEX, return_tensors='pt').unsqueeze(0).cuda()\n",
        "    stop_str = conv.sep if conv.sep_style != SeparatorStyle.TWO else conv.sep2\n",
        "    keywords = [stop_str]\n",
        "    stopping_criteria = KeywordsStoppingCriteria(keywords, tokenizer, input_ids)\n",
        "    with torch.inference_mode():\n",
        "      output_ids = model.generate(input_ids, images=image_tensor, do_sample=True, temperature=0.2, \n",
        "                                  max_new_tokens=1024, use_cache=True, stopping_criteria=[stopping_criteria])\n",
        "    outputs = tokenizer.decode(output_ids[0, input_ids.shape[1]:]).strip()\n",
        "    conv.messages[-1][-1] = outputs\n",
        "    output = outputs.rsplit('</s>', 1)[0]\n",
        "    return image, output"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {},
      "outputs": [],
      "source": [
        "image, output = caption_image(f'https://llava-vl.github.io/static/images/view.jpg', 'Describe the image and color details.')\n",
        "print(output)\n",
        "image"
      ]
    }
  ],
  "metadata": {
    "accelerator": "GPU",
    "colab": {
      "gpuType": "T4",
      "provenance": []
    },
    "kernelspec": {
      "display_name": "Python 3",
      "name": "python3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 0
}