{
 "cells": [
  {
   "cell_type": "code",
   "id": "initial_id",
   "metadata": {
    "collapsed": true,
    "ExecuteTime": {
     "end_time": "2024-12-14T14:40:23.089485Z",
     "start_time": "2024-12-14T14:40:22.937392Z"
    }
   },
   "source": [
    "import pandas as pd\n",
    "from src.utils.path_utils import get_project_root\n",
    "\n",
    "PROJECT_ROOT = get_project_root()"
   ],
   "outputs": [],
   "execution_count": 1
  },
  {
   "metadata": {
    "ExecuteTime": {
     "end_time": "2024-12-14T14:46:49.718444Z",
     "start_time": "2024-12-14T14:46:46.361765Z"
    }
   },
   "cell_type": "code",
   "source": [
    "import requests\n",
    "from PIL import Image\n",
    "from transformers import BlipProcessor, BlipForConditionalGeneration\n",
    "\n",
    "processor = BlipProcessor.from_pretrained(\"Salesforce/blip-image-captioning-large\")\n",
    "model = BlipForConditionalGeneration.from_pretrained(\"Salesforce/blip-image-captioning-large\")\n",
    "\n",
    "image = Image.open(f\"{PROJECT_ROOT}/data/scenery_image.jpg\")\n",
    "\n",
    "# conditional image captioning\n",
    "text = \"a photography of\"\n",
    "inputs = processor(image, text, return_tensors=\"pt\")\n",
    "\n",
    "out = model.generate(**inputs)\n",
    "print(processor.decode(out[0], skip_special_tokens=True))\n",
    "\n",
    "# unconditional image captioning\n",
    "inputs = processor(image, return_tensors=\"pt\")\n",
    "\n",
    "out = model.generate(**inputs)\n",
    "print(processor.decode(out[0], skip_special_tokens=True))\n"
   ],
   "id": "80b41a616dbbafd3",
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "a photography of a road leading to mountains with a sunset in the background\n",
      "arafed road with mountains in the background and a sunset\n"
     ]
    }
   ],
   "execution_count": 8
  },
  {
   "metadata": {},
   "cell_type": "code",
   "outputs": [],
   "execution_count": null,
   "source": "",
   "id": "983b19a8aa6e4a39"
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}