{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/mahnaz/mlprojects/bloom_classifier/ven_bloom_gradio/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "import gradio as gr\n",
    "import json\n",
    "from transformers import pipeline\n",
    "from transformers import AutoImageProcessor\n",
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "from torchvision.transforms import Compose, Resize, CenterCrop, ToTensor, Normalize\n",
    "import numpy as np\n",
    "\n",
    "def preprocess_input(input_data, image_processor):\n",
    "    \"\"\"\n",
    "    Preprocesses the input image for inference.\n",
    "\n",
    "    Parameters:\n",
    "        input_data (str or np.ndarray): Path to the image file in .jpg format or a NumPy array.\n",
    "        image_processor (AutoImageProcessor): An instance of AutoImageProcessor from the model's checkpoint.\n",
    "\n",
    "    Returns:\n",
    "        processed_img (torch.Tensor): Preprocessed image ready for inference.\n",
    "    \"\"\"\n",
    "    # Load the image based on the input type\n",
    "    if isinstance(input_data, str):\n",
    "        img = Image.open(input_data).convert('RGB')\n",
    "    elif isinstance(input_data, np.ndarray):\n",
    "        img = Image.fromarray(input_data.astype('uint8'), 'RGB')\n",
    "    else:\n",
    "        raise ValueError(\"Unsupported input type. Only str and np.ndarray are supported.\")\n",
    "    \n",
    "    # Obtain the mean and std from image_processor\n",
    "    mean = image_processor.image_mean\n",
    "    std = image_processor.image_std\n",
    "    \n",
    "    # Obtain the image size from image_processor\n",
    "    size = (\n",
    "        image_processor.size[\"shortest_edge\"]\n",
    "        if \"shortest_edge\" in image_processor.size\n",
    "        else (image_processor.size[\"height\"], image_processor.size[\"width\"])\n",
    "    )\n",
    "    \n",
    "    # Define the transformations\n",
    "    preprocess = Compose([\n",
    "        Resize(size),  # Resizing to the same size used during training\n",
    "        CenterCrop(size),  # Center cropping to the same size used during training\n",
    "        ToTensor(),\n",
    "        Normalize(mean=mean, std=std)\n",
    "    ])\n",
    "    \n",
    "    # Apply the transformations\n",
    "    processed_img = preprocess(img)\n",
    "    \n",
    "    # Add a batch dimension\n",
    "    processed_img = processed_img.unsqueeze(0)  # This is necessary because the model expects a batch\n",
    "       to_pil = ToPILImage()\n",
    "    processed_img = to_pil(processed_img)\n",
    "\n",
    "    return processed_img\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from PIL import Image\n",
    "from torchvision.transforms import Compose, Resize, CenterCrop, ToTensor, Normalize\n",
    "\n",
    "def preprocess_input(image_path, image_processor):\n",
    "    \"\"\"\n",
    "    Preprocesses the input image for inference.\n",
    "\n",
    "    Parameters:\n",
    "        image_path (str): Path to the image file in .jpg format.\n",
    "        image_processor (AutoImageProcessor): An instance of AutoImageProcessor from the model's checkpoint.\n",
    "\n",
    "    Returns:\n",
    "        processed_img (torch.Tensor): Preprocessed image ready for inference.\n",
    "    \"\"\"\n",
    "    # Load the image\n",
    "    img = Image.open(image_path).convert('RGB')\n",
    "    \n",
    "    # Obtain the mean and std from image_processor\n",
    "    mean = image_processor.image_mean\n",
    "    std = image_processor.image_std\n",
    "    \n",
    "    # Obtain the image size from image_processor\n",
    "    size = (\n",
    "        image_processor.size[\"shortest_edge\"]\n",
    "        if \"shortest_edge\" in image_processor.size\n",
    "        else (image_processor.size[\"height\"], image_processor.size[\"width\"])\n",
    "    )\n",
    "    \n",
    "    # Define the transformations\n",
    "    preprocess = Compose([\n",
    "        Resize(size),  # Resizing to the same size used during training\n",
    "        CenterCrop(size),  # Center cropping to the same size used during training\n",
    "        ToTensor(),\n",
    "        Normalize(mean=mean, std=std)\n",
    "    ])\n",
    "    \n",
    "    # Apply the transformations\n",
    "    processed_img = preprocess(img)\n",
    "    \n",
    "    # Add a batch dimension\n",
    "    processed_img = processed_img.unsqueeze(0)  # This is necessary because the model expects a batch\n",
    "\n",
    "    return processed_img\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/mahnaz/mlprojects/bloom_classifier/ven_bloom_gradio/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "import gradio as gr\n",
    "import json\n",
    "from transformers import pipeline\n",
    "\n",
    "\n",
    "def load_label_to_name_mapping(json_file_path):\n",
    "    \"\"\"Load the label-to-name mapping from a JSON file.\"\"\"\n",
    "    with open(json_file_path, 'r') as f:\n",
    "        mapping = json.load(f)\n",
    "    return {int(k): v for k, v in mapping.items()}\n",
    "\n",
    "def infer_flower_name(classifier, image):\n",
    "    \"\"\"Perform inference on an image and return the flower name.\"\"\"\n",
    "    # Perform inference\n",
    "    # Load the model checkpoint for inference\n",
    "    \n",
    "    result = classifier(image)\n",
    "    # Get the label from the inference result\n",
    "    label = result[0]['label'].split('_')[-1]  # The label is usually in the format 'LABEL_#'\n",
    "    label = int(label)\n",
    "    \n",
    "    # Map the integer label to the flower name\n",
    "    json_file_path = 'label_to_name.json'\n",
    "    label_to_name = load_label_to_name_mapping(json_file_path)\n",
    "    flower_name = label_to_name.get(label, \"Unknown\")\n",
    "    \n",
    "    return flower_name\n",
    "\n",
    "\n",
    "\n",
    "def predict(prompt_img):# would call a model to make a prediction on an input and return the output.\n",
    "\n",
    "    # Instantiate the AutoImageProcessor\n",
    "    #image_processor = AutoImageProcessor.from_pretrained(\"google/vit-base-patch16-224-in21k\")\n",
    "\n",
    "    # Preprocess the input image\n",
    "    #image_path = 'path/to/your/image.jpg'\n",
    "    #processed_img = preprocess_input(prompt_img, image_processor)\n",
    "    processed_img= prompt_img \n",
    "    classifier = pipeline(\"image-classification\", model=\"checkpoint-160\")\n",
    "    flower_name = infer_flower_name(classifier, processed_img)\n",
    "    return flower_name\n",
    "demo = gr.Interface(fn=predict, \n",
    "                    inputs=gr.Image(type=\"pil\"), \n",
    "                    outputs=gr.Label(num_top_classes=3),\n",
    "                    examples=[\"example.jpg\"])\n",
    "\n",
    "demo.launch()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv_bloom-classifier",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.3"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}