{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    },
    "language_info": {
      "name": "python"
    }
  },
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "metadata": {
        "id": "jaqWRbrAqXI2"
      },
      "outputs": [],
      "source": [
        "import gradio as gr\n",
        "import tensorflow as tf\n",
        "import numpy as np\n",
        "from PIL import Image\n",
        "\n",
        "# ขนาดภาพที่ใช้ในโมเดล\n",
        "IMG_SIZE = (224, 224)\n",
        "\n",
        "# สร้าง Dictionary ที่เก็บชื่อโมเดลและ path ไฟล์ .h5\n",
        "model_paths = {\n",
        "    \"Custom CNN\": \"Custom_CNN_model.h5\",\n",
        "    \"VGG16\": \"VGG16_model.h5\",\n",
        "    \"ResNet50\": \"ResNet50_model.h5\"\n",
        "}\n",
        "\n",
        "# ฟังก์ชันเตรียมข้อมูลภาพ\n",
        "def preprocess_image(image):\n",
        "    image = image.resize(IMG_SIZE)  # Resize\n",
        "    image = np.array(image) / 255.0  # Normalize\n",
        "    image = np.expand_dims(image, axis=0)  # เพิ่ม batch dimension\n",
        "    return image\n",
        "\n",
        "# ฟังก์ชันทำนาย โดยเลือกโมเดล\n",
        "def predict_with_model(image, model_name):\n",
        "    # โหลดโมเดลที่เลือก\n",
        "    model = tf.keras.models.load_model(model_paths[model_name])\n",
        "\n",
        "    # เตรียมภาพ\n",
        "    processed_image = preprocess_image(image)\n",
        "\n",
        "    # ทำนายผล\n",
        "    prediction = model.predict(processed_image)[0][0]  # ได้ค่าความน่าจะเป็น\n",
        "    class_name = \"Stroke\" if prediction > 0.5 else \"Non-Stroke\"\n",
        "    confidence = round(float(prediction if prediction > 0.5 else 1 - prediction) * 100, 2)\n",
        "\n",
        "    # คืนผลลัพธ์\n",
        "    return f\"Class: {class_name} (Confidence: {confidence}%)\"\n",
        "\n",
        "# Gradio Interface\n",
        "interface = gr.Interface(\n",
        "    fn=predict_with_model,\n",
        "    inputs=[\n",
        "        gr.Image(type=\"pil\", label=\"Upload Face Image\"),\n",
        "        gr.Dropdown(choices=[\"Custom CNN\", \"VGG16\", \"ResNet50\"], label=\"Select Model\")\n",
        "    ],\n",
        "    outputs=\"text\",\n",
        "    title=\"Stroke Face Classification\",\n",
        "    description=\"Upload a face image to predict whether the person has stroke or not. Select model to classify.\"\n",
        ")\n",
        "\n",
        "# Run app\n",
        "if __name__ == \"__main__\":\n",
        "    interface.launch()\n"
      ]
    }
  ]
}