{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 10,
   "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": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_path = \"Pokemon_transfer_learning.keras\"\n",
    "model = tf.keras.models.load_model(model_path)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define the core prediction function\n",
    "def predict_pokemon(image):\n",
    "    # Preprocess image\n",
    "    print(type(image))\n",
    "    image = Image.fromarray(image.astype('uint8'))  # Convert numpy array to PIL image\n",
    "    image = image.resize((150, 150)) #resize the image to 28x28 and converts it to gray scale\n",
    "    image = np.array(image)\n",
    "    image = np.expand_dims(image, axis=0) # same as image[None, ...]\n",
    "    \n",
    "    # Predict\n",
    "    prediction = model.predict(image)\n",
    "    \n",
    "    # Because the output layer was dense(0) without an activation function, we need to apply sigmoid to get the probability\n",
    "    # we could also change the output layer to dense(1, activation='sigmoid')\n",
    "    prediction = np.round(prediction, 2)\n",
    "        # Separate the probabilities for each class\n",
    "    p_abra = prediction[0][0]  # Probability for class 'abra'\n",
    "    p_beedrill = prediction[0][1]   # Probability for class 'moltres'\n",
    "    p_sandshrew = prediction[0][2]    # Probability for class 'zapdos'\n",
    "    return {'abra': p_abra, 'beedrill':  p_beedrill, 'sandshrew': p_sandshrew}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Running on local URL:  http://127.0.0.1:7862\n",
      "\n",
      "To create a public link, set `share=True` in `launch()`.\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div><iframe src=\"http://127.0.0.1:7862/\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": []
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'numpy.ndarray'>\n",
      "\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 1s/step\n"
     ]
    }
   ],
   "source": [
    "# Create the Gradio interface\n",
    "input_image = gr.Image()\n",
    "iface = gr.Interface(\n",
    "    fn=predict_pokemon,\n",
    "    inputs=input_image, \n",
    "    outputs=gr.Label(),\n",
    "    examples=[\"images/abra1.png\", \"images/abra2.jpg\", \"images/abra3.png\", \"images/beedrill1.png\", \"images/beedrill2.png\", \"images/beedrill3.jpg\", \"images/sandshrew1.png\", \"images/sandshrew2.jpg\", \"images/sandshrew3.png\"],   \n",
    "    description=\"A simple mlp classification model for image classification using the mnist dataset.\")\n",
    "iface.launch(share=True)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "venv_new",
   "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"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}