{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import gradio as gr\n",
    "import tensorflow as tf\n",
    "import numpy as np\n",
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Loading trained model\n",
    "model = tf.keras.models.load_model(\"model.h5\")  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Label Overview\n",
    "class_num = { 0:'Speed limit (20km/h)',\n",
    "            1:'Speed limit (30km/h)', +\n",
    "            2:'Speed limit (50km/h)', \n",
    "            3:'Speed limit (60km/h)', \n",
    "            4:'Speed limit (70km/h)', \n",
    "            5:'Speed limit (80km/h)', \n",
    "            6:'End of speed limit (80km/h)', \n",
    "            7:'Speed limit (100km/h)', \n",
    "            8:'Speed limit (120km/h)', \n",
    "            9:'No passing', \n",
    "            10:'No passing veh over 3.5 tons', \n",
    "            11:'Right-of-way at intersection', \n",
    "            12:'Priority road', \n",
    "            13:'Yield', \n",
    "            14:'Stop', \n",
    "            15:'No vehicles', \n",
    "            16:'Veh > 3.5 tons prohibited', \n",
    "            17:'No entry', \n",
    "            18:'General caution', \n",
    "            19:'Dangerous curve left', \n",
    "            20:'Dangerous curve right', \n",
    "            21:'Double curve', \n",
    "            22:'Bumpy road', \n",
    "            23:'Slippery road', \n",
    "            24:'Road narrows on the right', \n",
    "            25:'Road work', \n",
    "            26:'Traffic signals', \n",
    "            27:'Pedestrians', \n",
    "            28:'Children crossing', \n",
    "            29:'Bicycles crossing', \n",
    "            30:'Beware of ice/snow',\n",
    "            31:'Wild animals crossing', \n",
    "            32:'End speed + passing limits', \n",
    "            33:'Turn right ahead', \n",
    "            34:'Turn left ahead', \n",
    "            35:'Ahead only', \n",
    "            36:'Go straight or right', \n",
    "            37:'Go straight or left', \n",
    "            38:'Keep right', \n",
    "            39:'Keep left', \n",
    "            40:'Roundabout mandatory', \n",
    "            41:'End of no passing', \n",
    "            42:'End no passing veh > 3.5 tons' }\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "# Define a function to preprocess input image\n",
    "def preprocess_image(image):\n",
    "    # Resize the image to the required dimensions\n",
    "    image = image.resize((30, 30))\n",
    "    # Convert the PIL image to a NumPy array\n",
    "    image_array = np.array(image)\n",
    "    # Normalize pixel values to be between 0 and 1\n",
    "    image_array = image_array / 255.0\n",
    "    # Add batch dimension to the image\n",
    "    image_array = np.expand_dims(image_array, axis=0)\n",
    "    return image_array\n",
    "\n",
    "# Define a function to make predictions\n",
    "def predict_class(image):\n",
    "    # Preprocess the input image\n",
    "    processed_image = preprocess_image(image)\n",
    "    # Make predictions using the loaded model\n",
    "    predictions = model.predict(processed_image)\n",
    "    # Get the predicted class index\n",
    "    predicted_class_index = np.argmax(predictions)\n",
    "    # Return the predicted class label and class number\n",
    "    return f\"{class_num[predicted_class_index]} (Class {predicted_class_index})\"\n",
    "\n",
    "# Create Gradio interface\n",
    "iface = gr.Interface(\n",
    "    fn=predict_class,\n",
    "    inputs=gr.Image(type=\"pil\", label=\"Upload an image\"),\n",
    "    outputs=\"text\"\n",
    ")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Launch the Gradio interface\n",
    "iface.launch()\n"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}