From 8af733ff12040b626d5498bc9889b660d6a4e5ec Mon Sep 17 00:00:00 2001
From: rijalhabibullah <juangrizal@gmail.com>
Date: Sun, 19 Apr 2026 04:17:06 +0700
Subject: [PATCH] ganti api ke laravel

---
 api_client.py                                 | 262 ---------
 ...e_leaf_cnn_classification-checkpoint.ipynb | 372 ++++++------
 rice leaf diseases dataset/api_server.py      | 450 ---------------
 .../rice_leaf_cnn_classification.ipynb        | 530 +++++++-----------
 setup.bat                                     |  11 +-
 setup.sh                                      |  15 +-
 test_api.py                                   |  51 +-
 web_TA/.env.example                           |   4 +
 .../Controllers/ClassificationController.php  | 258 +++++++--
 .../Services/PythonClassificationService.php  | 130 +++++
 web_TA/routes/api.php                         |   7 +-
 web_TA/scripts/rice_inference.py              | 293 ++++++++++
 web_TA/~/.keras/keras.json                    |   6 +
 13 files changed, 1088 insertions(+), 1301 deletions(-)
 delete mode 100644 api_client.py
 delete mode 100644 rice leaf diseases dataset/api_server.py
 create mode 100644 web_TA/app/Services/PythonClassificationService.php
 create mode 100644 web_TA/scripts/rice_inference.py
 create mode 100644 web_TA/~/.keras/keras.json

diff --git a/api_client.py b/api_client.py
deleted file mode 100644
index 826f834..0000000
--- a/api_client.py
+++ /dev/null
@@ -1,262 +0,0 @@
-"""
-API Client Library - Python
-Untuk mengakses API Classification dari Python
-"""
-
-import requests
-import base64
-from pathlib import Path
-from typing import Dict, List, Optional, Tuple
-import json
-
-
-class RiceLeafClassificationAPI:
-    """Client untuk Rice Leaf Disease Classification API"""
-    
-    def __init__(self, base_url: str = "http://127.0.0.1:5000"):
-        """
-        Initialize API client
-        
-        Args:
-            base_url: Base URL untuk API endpoint (default: Flask API on port 5000)
-        """
-        self.base_url = base_url
-        self.session = requests.Session()
-        self.timeout = 30
-    
-    def test_connection(self) -> Tuple[bool, str]:
-        """
-        Test koneksi ke API
-        
-        Returns:
-            Tuple (success, message)
-        """
-        try:
-            response = self.session.get(f"{self.base_url}/health", timeout=10)
-            if response.status_code == 200:
-                data = response.json()
-                if 'status' in data:  # Flask API response
-                    return data['status'] == 'ok', data.get('message', 'Connection OK')
-                else:
-                    return data.get('success', True), data.get('message', 'Connection OK')
-            else:
-                return False, f"HTTP {response.status_code}"
-        except Exception as e:
-            return False, str(e)
-    
-    def classify_image(
-        self,
-        image_path: str,
-        save: bool = False,
-        notes: Optional[str] = None
-    ) -> Optional[Dict]:
-        """
-        Klasifikasi gambar
-        
-        Args:
-            image_path: Path ke file gambar
-            save: Jika True, simpan gambar ke server
-            notes: Catatan tambahan (hanya jika save=True)
-            
-        Returns:
-            Dict dengan hasil klasifikasi atau None jika error
-        """
-        # Validasi file
-        image_file = Path(image_path)
-        if not image_file.exists():
-            print(f"Error: File tidak ditemukan: {image_path}")
-            return None
-        
-        if not image_file.suffix.lower() in ['.jpg', '.jpeg', '.png', '.gif']:
-            print(f"Error: Format file tidak didukung: {image_file.suffix}")
-            return None
-        
-        try:
-            # Tentukan endpoint
-            endpoint = "classify" if not save else "classify"
-            url = f"{self.base_url}/{endpoint}"
-            
-            # Baca dan kirim file
-            with open(image_file, 'rb') as f:
-                files = {'image': f}
-                data = {}
-                if save and notes:
-                    data['notes'] = notes
-                
-                response = self.session.post(
-                    url,
-                    files=files,
-                    data=data,
-                    timeout=self.timeout
-                )
-            
-            if response.status_code == 200:
-                response_data = response.json()
-                if response_data.get('success'):
-                    return response_data.get('data')
-                else:
-                    print(f"Error: {response_data.get('message')}")
-                    return None
-            else:
-                print(f"Error: HTTP {response.status_code}")
-                return None
-                
-        except Exception as e:
-            print(f"Error during classification: {str(e)}")
-            return None
-    
-    def classify_from_base64(
-        self,
-        base64_image: str,
-        filename: str = "image.jpg"
-    ) -> Optional[Dict]:
-        """
-        Klasifikasi dari base64 string
-        
-        Args:
-            base64_image: Base64 encoded image string
-            filename: Nama file (opsional)
-            
-        Returns:
-            Dict dengan hasil klasifikasi
-        """
-        try:
-            url = f"{self.base_url}/classify"
-            payload = {
-                "image": base64_image,
-                "filename": filename
-            }
-            
-            response = self.session.post(
-                url,
-                json=payload,
-                timeout=self.timeout
-            )
-            
-            if response.status_code == 200:
-                response_data = response.json()
-                if response_data.get('success'):
-                    return response_data.get('data')
-                else:
-                    print(f"Error: {response_data.get('message')}")
-                    return None
-            else:
-                print(f"Error: HTTP {response.status_code}")
-                return None
-                
-        except Exception as e:
-            print(f"Error: {str(e)}")
-            return None
-    
-    def batch_classify(
-        self,
-        image_paths: List[str],
-        save: bool = False
-    ) -> List[Dict]:
-        """
-        Klasifikasi multiple gambar sekaligus
-        
-        Args:
-            image_paths: List path-ke-gambar
-            save: Simpan ke server
-            
-        Returns:
-            List hasil klasifikasi
-        """
-        results = []
-        total = len(image_paths)
-        
-        for idx, path in enumerate(image_paths, 1):
-            print(f"Processing {idx}/{total}: {Path(path).name}...", end=" ")
-            result = self.classify_image(path, save=save)
-            
-            if result:
-                print("✓")
-                results.append({
-                    'image': path,
-                    'result': result
-                })
-            else:
-                print("✗")
-                results.append({
-                    'image': path,
-                    'result': None
-                })
-        
-        return results
-    
-    def print_result(self, result: Dict):
-        """Print hasil klasifikasi dalam format yang dapat dibaca"""
-        print("\n" + "="*60)
-        print("KLASIFIKASI HASIL")
-        print("="*60)
-        
-        print(f"\n🎯 DIAGNOSIS: {result['disease_info']['name']}")
-        print(f"   Confidence: {result['confidence']}")
-        print(f"   Severity: {result['disease_info']['severity']}")
-        
-        print(f"\n📊 PREDIKSI SEMUA KELAS:")
-        for class_name, score in result['all_predictions'].items():
-            percentage = f"{score*100:.2f}%"
-            bar = "█" * int(score * 20)
-            print(f"   {class_name:20} {percentage:>8} {bar}")
-        
-        print(f"\n🔬 GEJALA:")
-        for symptom in result['disease_info']['symptoms']:
-            print(f"   • {symptom}")
-        
-        print(f"\n💊 PENANGANAN:")
-        for treatment in result['disease_info']['treatment']:
-            print(f"   • {treatment}")
-        
-        print("\n" + "="*60 + "\n")
-
-
-# ============================================================================
-# CONTOH PENGGUNAAN
-# ============================================================================
-
-if __name__ == "__main__":
-    # Initialize client
-    api = RiceLeafClassificationAPI()
-    
-    # Test connection
-    print("Testing API connection...")
-    success, message = api.test_connection()
-    if success:
-        print(f"✓ {message}\n")
-    else:
-        print(f"✗ Failed: {message}\n")
-        exit(1)
-    
-    # Contoh 1: Klasifikasi single image
-    print("Example 1: Klasifikasi single image")
-    print("-" * 60)
-    image_path = "path/to/rice_leaf.jpg"  # Ganti dengan path asli
-    result = api.classify_image(image_path, save=True, notes="Test dari script")
-    if result:
-        api.print_result(result)
-    
-    # Contoh 2: Batch classification
-    print("\nExample 2: Batch classification")
-    print("-" * 60)
-    image_list = [
-        "path/to/image1.jpg",
-        "path/to/image2.jpg",
-        "path/to/image3.jpg",
-    ]
-    results = api.batch_classify(image_list, save=True)
-    
-    # Summary
-    successful = sum(1 for r in results if r['result'] is not None)
-    print(f"\nBatch Summary: {successful}/{len(results)} berhasil")
-    
-    # Contoh 3: Klasifikasi dari base64
-    print("\nExample 3: Klasifikasi dari base64")
-    print("-" * 60)
-    with open("image.jpg", "rb") as f:
-        base64_image = base64.b64encode(f.read()).decode('utf-8')
-    
-    result = api.classify_from_base64(base64_image)
-    if result:
-        api.print_result(result)
diff --git a/rice leaf diseases dataset/.ipynb_checkpoints/rice_leaf_cnn_classification-checkpoint.ipynb b/rice leaf diseases dataset/.ipynb_checkpoints/rice_leaf_cnn_classification-checkpoint.ipynb
index b09054c..c4d8709 100644
--- a/rice leaf diseases dataset/.ipynb_checkpoints/rice_leaf_cnn_classification-checkpoint.ipynb	
+++ b/rice leaf diseases dataset/.ipynb_checkpoints/rice_leaf_cnn_classification-checkpoint.ipynb	
@@ -2,16 +2,19 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 4,
    "id": "5ca735e6",
    "metadata": {},
    "outputs": [
     {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "TensorFlow Version: 2.20.0\n",
-      "GPU Available: []\n"
+     "ename": "ModuleNotFoundError",
+     "evalue": "No module named 'pandas'",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[1;31mModuleNotFoundError\u001b[0m                       Traceback (most recent call last)",
+      "Cell \u001b[1;32mIn[4], line 3\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mos\u001b[39;00m\n\u001b[0;32m      2\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mnumpy\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mnp\u001b[39;00m\n\u001b[1;32m----> 3\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mpandas\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mpd\u001b[39;00m\n\u001b[0;32m      4\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mmatplotlib\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpyplot\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mplt\u001b[39;00m\n\u001b[0;32m      5\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mseaborn\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01msns\u001b[39;00m\n",
+      "\u001b[1;31mModuleNotFoundError\u001b[0m: No module named 'pandas'"
      ]
     }
    ],
@@ -69,7 +72,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": null,
    "id": "63695071",
    "metadata": {},
    "outputs": [
@@ -234,6 +237,9 @@
    ],
    "source": [
     "# Set dataset path - handle both relative and absolute paths\n",
+    "from pathlib import Path\n",
+    "import re\n",
+    "\n",
     "notebook_dir = Path('.').absolute()\n",
     "print(f\"Notebook working directory: {notebook_dir}\\n\")\n",
     "\n",
@@ -258,47 +264,75 @@
     "\n",
     "classes = ['Bacterialblight', 'Brownspot', 'Leafsmut']\n",
     "\n",
-    "# Load images and labels\n",
+    "# Heuristic filter to remove offline augmented/duplicate files.\n",
+    "def is_augmented_or_duplicate(filename: str) -> bool:\n",
+    "    name = filename.lower()\n",
+    "    patterns = [\n",
+    "        r'rotated',\n",
+    "        r'\\borig\\b',\n",
+    "        r'_orig_',\n",
+    "        r'\\(\\d+\\)',\n",
+    "        r'copy',\n",
+    "    ]\n",
+    "    return any(re.search(p, name) for p in patterns)\n",
+    "\n",
+    "# Build group id so near-duplicate name variants do not leak across splits.\n",
+    "def build_group_id(class_name: str, filename: str) -> str:\n",
+    "    stem = Path(filename).stem.lower()\n",
+    "    stem = re.sub(r'\\(\\d+\\)', '', stem)\n",
+    "    stem = re.sub(r'\\s+', '_', stem)\n",
+    "    stem = re.sub(r'_+', '_', stem).strip('_')\n",
+    "    return f\"{class_name}:{stem}\"\n",
+    "\n",
+    "# Load images, labels, and group ids\n",
     "images = []\n",
     "labels = []\n",
+    "groups = []\n",
     "class_to_idx = {class_name: idx for idx, class_name in enumerate(classes)}\n",
     "\n",
-    "print(\"Loading dataset...\")\n",
+    "print(\"Loading dataset (with anti-leakage filtering)...\")\n",
     "print(\"-\" * 50)\n",
     "\n",
     "total_images = 0\n",
+    "total_filtered_out = 0\n",
+    "\n",
     "for class_name in classes:\n",
     "    class_path = dataset_base_path / class_name\n",
-    "    \n",
+    "\n",
     "    if class_path.exists():\n",
-    "        # Count images first for progress bar\n",
     "        image_files = [f for f in class_path.glob('*') if f.suffix.lower() in ['.jpg', '.jpeg', '.png']]\n",
-    "        \n",
-    "        # Load images with progress bar\n",
+    "        kept_files = [f for f in image_files if not is_augmented_or_duplicate(f.name)]\n",
+    "        filtered_count = len(image_files) - len(kept_files)\n",
+    "\n",
     "        image_count = 0\n",
-    "        with tqdm(image_files, desc=f\"Loading {class_name}\", position=classes.index(class_name), leave=True) as pbar:\n",
+    "        with tqdm(kept_files, desc=f\"Loading {class_name}\", position=classes.index(class_name), leave=True) as pbar:\n",
     "            for img_file in pbar:\n",
     "                img = cv2.imread(str(img_file))\n",
     "                if img is not None:\n",
     "                    img_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)\n",
     "                    images.append(img_rgb)\n",
     "                    labels.append(class_to_idx[class_name])\n",
+    "                    groups.append(build_group_id(class_name, img_file.name))\n",
     "                    image_count += 1\n",
-    "        \n",
-    "        print(f\"  ✓ {class_name}: {image_count} images loaded\\n\")\n",
+    "\n",
+    "        print(f\"  ✓ {class_name}: {image_count} images loaded | filtered out: {filtered_count}\\n\")\n",
     "        total_images += image_count\n",
+    "        total_filtered_out += filtered_count\n",
     "    else:\n",
     "        print(f\"  ✗ {class_name} folder not found!\\n\")\n",
     "\n",
     "print(\"-\" * 50)\n",
-    "print(f\"Total images loaded: {total_images}\\n\")\n",
+    "print(f\"Total images loaded: {total_images}\")\n",
+    "print(f\"Total files filtered out: {total_filtered_out}\\n\")\n",
     "\n",
     "# Convert to numpy arrays with proper dtype\n",
     "X = np.array(images)\n",
-    "y = np.array(labels, dtype=np.int32)  # Convert to int32 for bincount\n",
+    "y = np.array(labels, dtype=np.int32)\n",
+    "group_ids = np.array(groups)\n",
     "\n",
     "print(f\"Dataset shape: {X.shape}\")\n",
     "print(f\"Labels shape: {y.shape}\")\n",
+    "print(f\"Unique group ids: {len(np.unique(group_ids))}\")\n",
     "if len(y) > 0:\n",
     "    print(f\"Class distribution: {np.bincount(y)}\")\n",
     "else:\n",
@@ -307,7 +341,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": null,
    "id": "a9a6d821",
    "metadata": {},
    "outputs": [
@@ -364,7 +398,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "id": "f8de0b07",
    "metadata": {},
    "outputs": [
@@ -416,35 +450,61 @@
     "        resized_img = cv2.resize(img, (IMG_SIZE, IMG_SIZE))\n",
     "        X_resized.append(resized_img)\n",
     "\n",
-    "X_resized = np.array(X_resized)\n",
+    "X_resized = np.array(X_resized).astype('float32')\n",
     "\n",
-    "# Normalize pixel values to [0, 1]\n",
-    "print(\"Normalizing pixel values...\")\n",
-    "X_normalized = X_resized.astype('float32') / 255.0\n",
+    "# MobileNetV2-specific preprocessing ([-1, 1] range)\n",
+    "from tensorflow.keras.applications.mobilenet_v2 import preprocess_input\n",
+    "print(\"Applying MobileNetV2 preprocess_input...\")\n",
+    "X_preprocessed = preprocess_input(X_resized)\n",
     "\n",
-    "print(f\"\\nPreprocessed image shape: {X_normalized.shape}\")\n",
-    "print(f\"Pixel value range: [{X_normalized.min()}, {X_normalized.max()}]\")\n",
+    "print(f\"\\nPreprocessed image shape: {X_preprocessed.shape}\")\n",
+    "print(f\"Pixel value range: [{X_preprocessed.min():.3f}, {X_preprocessed.max():.3f}]\")\n",
     "\n",
-    "# Split dataset into train, validation, and test sets\n",
-    "print(\"\\nSplitting dataset...\")\n",
-    "X_train, X_temp, y_train, y_temp = train_test_split(\n",
-    "    X_normalized, y, test_size=0.3, random_state=42, stratify=y\n",
-    ")\n",
+    "# Split dataset into train, validation, and test sets (group-aware anti-leakage split)\n",
+    "print(\"\\nSplitting dataset with GROUP-AWARE strategy (anti data leakage)...\")\n",
+    "from sklearn.model_selection import GroupShuffleSplit\n",
     "\n",
-    "X_val, X_test, y_val, y_test = train_test_split(\n",
-    "    X_temp, y_temp, test_size=0.5, random_state=42, stratify=y_temp\n",
-    ")\n",
+    "# 70% train, 30% temp (val+test)\n",
+    "gss_outer = GroupShuffleSplit(n_splits=1, test_size=0.30, random_state=42)\n",
+    "train_idx, temp_idx = next(gss_outer.split(X_preprocessed, y, groups=group_ids))\n",
+    "\n",
+    "X_train, X_temp = X_preprocessed[train_idx], X_preprocessed[temp_idx]\n",
+    "y_train, y_temp = y[train_idx], y[temp_idx]\n",
+    "groups_train, groups_temp = group_ids[train_idx], group_ids[temp_idx]\n",
+    "\n",
+    "# Split temp equally into val and test: 15% val, 15% test\n",
+    "gss_inner = GroupShuffleSplit(n_splits=1, test_size=0.50, random_state=42)\n",
+    "val_rel_idx, test_rel_idx = next(gss_inner.split(X_temp, y_temp, groups=groups_temp))\n",
+    "\n",
+    "X_val, X_test = X_temp[val_rel_idx], X_temp[test_rel_idx]\n",
+    "y_val, y_test = y_temp[val_rel_idx], y_temp[test_rel_idx]\n",
+    "groups_val, groups_test = groups_temp[val_rel_idx], groups_temp[test_rel_idx]\n",
+    "\n",
+    "# Leakage sanity check: these must all be zero\n",
+    "leak_train_val = len(set(groups_train) & set(groups_val))\n",
+    "leak_train_test = len(set(groups_train) & set(groups_test))\n",
+    "leak_val_test = len(set(groups_val) & set(groups_test))\n",
     "\n",
     "print(\"-\" * 50)\n",
-    "print(f\"Training set: {X_train.shape[0]} images\")\n",
+    "print(f\"Training set:   {X_train.shape[0]} images\")\n",
     "print(f\"Validation set: {X_val.shape[0]} images\")\n",
-    "print(f\"Test set: {X_test.shape[0]} images\")\n",
-    "print(\"-\" * 50)"
+    "print(f\"Test set:       {X_test.shape[0]} images\")\n",
+    "print(\"-\" * 50)\n",
+    "print(\"Leakage check (should be 0):\")\n",
+    "print(f\"  Train ∩ Val : {leak_train_val}\")\n",
+    "print(f\"  Train ∩ Test: {leak_train_test}\")\n",
+    "print(f\"  Val ∩ Test  : {leak_val_test}\")\n",
+    "print(\"-\" * 50)\n",
+    "\n",
+    "# Show class distribution after split\n",
+    "for split_name, split_y in [('Train', y_train), ('Val', y_val), ('Test', y_test)]:\n",
+    "    counts = np.bincount(split_y, minlength=len(classes))\n",
+    "    print(f\"{split_name} distribution: {counts}\")"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": null,
    "id": "eba217e2",
    "metadata": {},
    "outputs": [
@@ -465,22 +525,21 @@
     }
    ],
    "source": [
-    "# Data augmentation\n",
+    "# Data augmentation (moderate, realistic)\n",
     "print(\"Configuring data augmentation...\")\n",
     "print(\"-\" * 50)\n",
     "\n",
     "train_datagen = ImageDataGenerator(\n",
-    "    rotation_range=20,\n",
-    "    width_shift_range=0.2,\n",
-    "    height_shift_range=0.2,\n",
+    "    rotation_range=15,\n",
+    "    width_shift_range=0.10,\n",
+    "    height_shift_range=0.10,\n",
+    "    zoom_range=0.10,\n",
     "    horizontal_flip=True,\n",
-    "    vertical_flip=True,\n",
-    "    zoom_range=0.2,\n",
-    "    shear_range=0.2,\n",
     "    fill_mode='nearest'\n",
     ")\n",
     "\n",
-    "val_datagen = ImageDataGenerator()  # No augmentation for validation\n",
+    "# No augmentation for validation/test\n",
+    "val_datagen = ImageDataGenerator()\n",
     "\n",
     "# Convert to one-hot encoding\n",
     "print(\"Converting labels to one-hot encoding...\")\n",
@@ -506,7 +565,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "id": "146d5d58",
    "metadata": {},
    "outputs": [
@@ -756,69 +815,30 @@
     }
    ],
    "source": [
-    "# Option 1: Build custom CNN model from scratch\n",
-    "print(\"Building custom CNN model...\")\n",
+    "# Build Transfer Learning model (recommended to reduce overfitting)\n",
+    "print(\"Building MobileNetV2 transfer learning model...\")\n",
     "print(\"-\" * 50)\n",
-    "print(\"Architecture:\")\n",
-    "print(\"  - 4 Convolutional Blocks (32→64→128→256 filters)\")\n",
-    "print(\"  - Batch Normalization & Dropout for regularization\")\n",
-    "print(\"  - Global Average Pooling + Dense layers\")\n",
-    "print()\n",
     "\n",
-    "def build_custom_cnn():\n",
-    "    model = models.Sequential([\n",
-    "        # Block 1\n",
-    "        layers.Conv2D(32, (3, 3), activation='relu', padding='same', input_shape=(IMG_SIZE, IMG_SIZE, 3)),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.Conv2D(32, (3, 3), activation='relu', padding='same'),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.MaxPooling2D((2, 2)),\n",
-    "        layers.Dropout(0.25),\n",
-    "        \n",
-    "        # Block 2\n",
-    "        layers.Conv2D(64, (3, 3), activation='relu', padding='same'),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.Conv2D(64, (3, 3), activation='relu', padding='same'),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.MaxPooling2D((2, 2)),\n",
-    "        layers.Dropout(0.25),\n",
-    "        \n",
-    "        # Block 3\n",
-    "        layers.Conv2D(128, (3, 3), activation='relu', padding='same'),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.Conv2D(128, (3, 3), activation='relu', padding='same'),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.MaxPooling2D((2, 2)),\n",
-    "        layers.Dropout(0.25),\n",
-    "        \n",
-    "        # Block 4\n",
-    "        layers.Conv2D(256, (3, 3), activation='relu', padding='same'),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.Conv2D(256, (3, 3), activation='relu', padding='same'),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.MaxPooling2D((2, 2)),\n",
-    "        layers.Dropout(0.25),\n",
-    "        \n",
-    "        # Global Average Pooling\n",
-    "        layers.GlobalAveragePooling2D(),\n",
-    "        \n",
-    "        # Dense layers\n",
-    "        layers.Dense(512, activation='relu'),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.Dropout(0.5),\n",
-    "        \n",
-    "        layers.Dense(256, activation='relu'),\n",
-    "        layers.BatchNormalization(),\n",
-    "        layers.Dropout(0.5),\n",
-    "        \n",
-    "        # Output layer\n",
-    "        layers.Dense(len(classes), activation='softmax')\n",
-    "    ])\n",
-    "    \n",
-    "    return model\n",
+    "from tensorflow.keras.applications import MobileNetV2\n",
+    "from tensorflow.keras import regularizers\n",
+    "\n",
+    "base_model = MobileNetV2(\n",
+    "    input_shape=(IMG_SIZE, IMG_SIZE, 3),\n",
+    "    include_top=False,\n",
+    "    weights='imagenet'\n",
+    ")\n",
+    "base_model.trainable = False  # freeze backbone at first stage\n",
+    "\n",
+    "model = models.Sequential([\n",
+    "    base_model,\n",
+    "    layers.GlobalAveragePooling2D(),\n",
+    "    layers.Dropout(0.4),\n",
+    "    layers.Dense(128, activation='relu', kernel_regularizer=regularizers.l2(1e-4)),\n",
+    "    layers.BatchNormalization(),\n",
+    "    layers.Dropout(0.4),\n",
+    "    layers.Dense(len(classes), activation='softmax')\n",
+    "])\n",
     "\n",
-    "print(\"Creating model...\")\n",
-    "model = build_custom_cnn()\n",
     "print(\"✓ Model created!\\n\")\n",
     "print(\"Model Summary:\")\n",
     "print(\"-\" * 50)\n",
@@ -828,7 +848,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": null,
    "id": "240316b9",
    "metadata": {},
    "outputs": [
@@ -849,30 +869,18 @@
     }
    ],
    "source": [
-    "# Option 2: Transfer Learning with MobileNetV2 (uncomment to use)\n",
-    "# base_model = MobileNetV2(input_shape=(IMG_SIZE, IMG_SIZE, 3), include_top=False, weights='imagenet')\n",
-    "# base_model.trainable = False\n",
-    "# \n",
-    "# model = models.Sequential([\n",
-    "#     base_model,\n",
-    "#     layers.GlobalAveragePooling2D(),\n",
-    "#     layers.Dense(256, activation='relu'),\n",
-    "#     layers.Dropout(0.5),\n",
-    "#     layers.Dense(len(classes), activation='softmax')\n",
-    "# ])\n",
-    "\n",
-    "# Compile the model\n",
+    "# Compile the model with stronger regularization\n",
     "print(\"Compiling model...\")\n",
     "print(\"-\" * 50)\n",
     "model.compile(\n",
-    "    optimizer=keras.optimizers.Adam(learning_rate=0.001),\n",
-    "    loss='categorical_crossentropy',\n",
+    "    optimizer=keras.optimizers.Adam(learning_rate=1e-4),\n",
+    "    loss=keras.losses.CategoricalCrossentropy(label_smoothing=0.1),\n",
     "    metrics=['accuracy']\n",
     ")\n",
     "\n",
     "print(\"Configuration:\")\n",
-    "print(\"  - Optimizer: Adam (lr=0.001)\")\n",
-    "print(\"  - Loss: Categorical Crossentropy\")\n",
+    "print(\"  - Optimizer: Adam (lr=0.0001)\")\n",
+    "print(\"  - Loss: Categorical Crossentropy + Label Smoothing (0.1)\")\n",
     "print(\"  - Metrics: Accuracy\")\n",
     "print(\"-\" * 50)\n",
     "print(\"✓ Model compiled successfully!\\n\")"
@@ -888,7 +896,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": null,
    "id": "26699eff",
    "metadata": {},
    "outputs": [
@@ -967,78 +975,126 @@
    "source": [
     "# Define callbacks with custom callback for progress display\n",
     "class ProgressCallback(keras.callbacks.Callback):\n",
-    "    def __init__(self):\n",
+    "    def __init__(self, total_epochs):\n",
     "        self.start_time = None\n",
     "        self.epoch_times = []\n",
-    "    \n",
+    "        self.total_epochs = total_epochs\n",
+    "\n",
     "    def on_train_begin(self, logs=None):\n",
     "        self.start_time = time.time()\n",
     "        self.epoch_times = []\n",
     "        print(\"\\n📊 Training Progress:\")\n",
     "        print(\"-\" * 80)\n",
-    "    \n",
+    "\n",
     "    def on_epoch_end(self, epoch, logs=None):\n",
     "        epoch_time = time.time() - self.start_time\n",
     "        self.epoch_times.append(epoch_time)\n",
     "        avg_epoch_time = epoch_time / (epoch + 1)\n",
-    "        remaining_epochs = EPOCHS - (epoch + 1)\n",
-    "        eta_seconds = avg_epoch_time * remaining_epochs\n",
-    "        \n",
+    "        remaining_epochs = self.total_epochs - (epoch + 1)\n",
+    "        eta_seconds = avg_epoch_time * max(remaining_epochs, 0)\n",
+    "\n",
     "        if logs:\n",
     "            loss = logs.get('loss', 0)\n",
     "            val_loss = logs.get('val_loss', 0)\n",
     "            acc = logs.get('accuracy', 0)\n",
     "            val_acc = logs.get('val_accuracy', 0)\n",
-    "            \n",
-    "            progress = \"█\" * (epoch + 1) + \"░\" * (EPOCHS - epoch - 1)\n",
-    "            print(f\"[{progress}] Epoch {epoch + 1}/{EPOCHS} | \"\n",
+    "\n",
+    "            progress = \"█\" * (epoch + 1) + \"░\" * (self.total_epochs - epoch - 1)\n",
+    "            print(f\"[{progress}] Epoch {epoch + 1}/{self.total_epochs} | \"\n",
     "                  f\"Loss: {loss:.4f} | Val Loss: {val_loss:.4f} | \"\n",
     "                  f\"Acc: {acc:.4f} | Val Acc: {val_acc:.4f} | \"\n",
     "                  f\"ETA: {int(eta_seconds)}s\")\n",
-    "    \n",
+    "\n",
     "    def on_train_end(self, logs=None):\n",
     "        total_time = time.time() - self.start_time\n",
     "        print(\"-\" * 80)\n",
     "        print(f\"✓ Training completed in {int(total_time)}s\\n\")\n",
     "\n",
     "import time\n",
+    "import numpy as np\n",
+    "from sklearn.utils.class_weight import compute_class_weight\n",
     "\n",
+    "# Balanced class weights to reduce bias to dominant class\n",
+    "class_weights = compute_class_weight(\n",
+    "    class_weight='balanced',\n",
+    "    classes=np.unique(y_train),\n",
+    "    y=y_train\n",
+    ")\n",
+    "class_weight_dict = {i: w for i, w in enumerate(class_weights)}\n",
+    "print(\"Class weights:\", class_weight_dict)\n",
+    "\n",
+    "# Stage 1 callbacks\n",
     "early_stop = keras.callbacks.EarlyStopping(\n",
     "    monitor='val_loss',\n",
-    "    patience=10,\n",
+    "    patience=6,\n",
     "    restore_best_weights=True\n",
     ")\n",
     "\n",
     "reduce_lr = keras.callbacks.ReduceLROnPlateau(\n",
     "    monitor='val_loss',\n",
     "    factor=0.5,\n",
-    "    patience=5,\n",
-    "    min_lr=1e-7\n",
+    "    patience=3,\n",
+    "    min_lr=1e-7,\n",
+    "    verbose=1\n",
     ")\n",
     "\n",
-    "# Train the model\n",
-    "EPOCHS = 50\n",
+    "checkpoint = keras.callbacks.ModelCheckpoint(\n",
+    "    'best_rice_leaf_model.keras',\n",
+    "    monitor='val_loss',\n",
+    "    save_best_only=True,\n",
+    "    verbose=1\n",
+    ")\n",
+    "\n",
+    "# Stage 1: train head only\n",
+    "EPOCHS_STAGE1 = 20\n",
     "BATCH_SIZE = 32\n",
     "\n",
     "print(\"=\" * 80)\n",
-    "print(\"🚀 STARTING MODEL TRAINING\")\n",
+    "print(\"🚀 STAGE 1 TRAINING (Frozen Backbone)\")\n",
     "print(\"=\" * 80)\n",
-    "print(f\"Epochs: {EPOCHS} | Batch Size: {BATCH_SIZE}\")\n",
+    "print(f\"Epochs: {EPOCHS_STAGE1} | Batch Size: {BATCH_SIZE}\")\n",
     "print(f\"Training samples: {len(X_train)} | Validation samples: {len(X_val)}\")\n",
-    "print(f\"Optimizer: Adam (lr=0.001)\")\n",
-    "print(f\"Loss Function: Categorical Crossentropy\")\n",
+    "print(f\"Optimizer: Adam (lr=0.0001)\")\n",
     "print(\"=\" * 80)\n",
     "\n",
-    "history = model.fit(\n",
+    "history_stage1 = model.fit(\n",
     "    train_datagen.flow(X_train, y_train_cat, batch_size=BATCH_SIZE),\n",
-    "    epochs=EPOCHS,\n",
-    "    batch_size=BATCH_SIZE,\n",
+    "    epochs=EPOCHS_STAGE1,\n",
     "    validation_data=(X_val, y_val_cat),\n",
-    "    callbacks=[early_stop, reduce_lr, ProgressCallback()],\n",
-    "    verbose=0  # Suppress default verbose output\n",
+    "    class_weight=class_weight_dict,\n",
+    "    callbacks=[early_stop, reduce_lr, checkpoint, ProgressCallback(EPOCHS_STAGE1)],\n",
+    "    verbose=0\n",
     ")\n",
     "\n",
-    "print(\"✅ Model training and validation completed!\")"
+    "# Stage 2: fine-tune last layers of MobileNetV2\n",
+    "print(\"\\n\" + \"=\" * 80)\n",
+    "print(\"🔧 STAGE 2 FINE-TUNING (Unfreeze top layers)\")\n",
+    "print(\"=\" * 80)\n",
+    "\n",
+    "base_model.trainable = True\n",
+    "for layer in base_model.layers[:-40]:\n",
+    "    layer.trainable = False\n",
+    "\n",
+    "model.compile(\n",
+    "    optimizer=keras.optimizers.Adam(learning_rate=1e-5),\n",
+    "    loss=keras.losses.CategoricalCrossentropy(label_smoothing=0.1),\n",
+    "    metrics=['accuracy']\n",
+    ")\n",
+    "\n",
+    "EPOCHS_STAGE2 = 10\n",
+    "print(f\"Fine-tuning epochs: {EPOCHS_STAGE2} | lr=1e-5\")\n",
+    "\n",
+    "history_stage2 = model.fit(\n",
+    "    train_datagen.flow(X_train, y_train_cat, batch_size=BATCH_SIZE),\n",
+    "    epochs=EPOCHS_STAGE2,\n",
+    "    validation_data=(X_val, y_val_cat),\n",
+    "    class_weight=class_weight_dict,\n",
+    "    callbacks=[early_stop, reduce_lr, checkpoint, ProgressCallback(EPOCHS_STAGE2)],\n",
+    "    verbose=0\n",
+    ")\n",
+    "\n",
+    "print(\"✅ Training + fine-tuning completed!\")\n",
+    "print(\"Best model saved to: best_rice_leaf_model.keras\")"
    ]
   },
   {
@@ -1051,7 +1107,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
    "id": "0c34bf89",
    "metadata": {},
    "outputs": [
@@ -1136,7 +1192,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": null,
    "id": "586213e5",
    "metadata": {},
    "outputs": [
@@ -1232,7 +1288,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": null,
    "id": "a0bffc2c",
    "metadata": {},
    "outputs": [
@@ -1347,7 +1403,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": null,
    "id": "ac7f14c3",
    "metadata": {},
    "outputs": [
@@ -1483,7 +1539,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": null,
    "id": "783a7e43",
    "metadata": {},
    "outputs": [
@@ -1674,7 +1730,7 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "base",
    "language": "python",
    "name": "python3"
   },
@@ -1688,7 +1744,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.12.5"
   }
  },
  "nbformat": 4,
diff --git a/rice leaf diseases dataset/api_server.py b/rice leaf diseases dataset/api_server.py
deleted file mode 100644
index 26eea5f..0000000
--- a/rice leaf diseases dataset/api_server.py	
+++ /dev/null
@@ -1,450 +0,0 @@
-"""
-Flask API Server untuk Rice Leaf Disease Classification
-Menjalankan model TensorFlow dan melayani request dari Laravel
-"""
-
-import os
-import json
-import base64
-import numpy as np
-from PIL import Image
-from io import BytesIO
-from pathlib import Path
-import warnings
-warnings.filterwarnings('ignore')
-
-# TensorFlow & Keras
-try:
-    import tensorflow as tf
-    from tensorflow import keras
-    from tensorflow.keras.preprocessing.image import img_to_array
-    TENSORFLOW_AVAILABLE = True
-except ImportError as e:
-    print(f"⚠️  Warning: TensorFlow not available: {e}")
-    print("   Using mock predictions for testing")
-    TENSORFLOW_AVAILABLE = False
-    keras = None
-
-# Flask
-from flask import Flask, request, jsonify
-from flask_cors import CORS
-
-# Inisialisasi Flask
-app = Flask(__name__)
-CORS(app)
-
-# Konfigurasi
-MODEL_PATH = None  # Akan diset saat startup
-MODEL = None
-IMG_SIZE = (224, 224)
-CLASS_NAMES = ['Bacterialblight', 'Brownspot', 'Leafsmut']
-
-def load_model():
-    """Load model dari file yang tersedia"""
-    global MODEL_PATH, MODEL
-    
-    if not TENSORFLOW_AVAILABLE:
-        print("⚠️  TensorFlow not available - Using mock mode for testing")
-        MODEL_PATH = "MOCK_MODEL"
-        return True
-    
-    # Cari model file yang tersedia
-    possible_models = [
-        'rice_leaf_disease_model.keras',
-        'rice_leaf_disease_model.h5',
-        'rice_leaf_disease_model.json'
-    ]
-    
-    for model_name in possible_models:
-        if os.path.exists(model_name):
-            MODEL_PATH = model_name
-            print(f"✓ Model ditemukan: {MODEL_PATH}")
-            
-            try:
-                if model_name.endswith('.keras'):
-                    MODEL = keras.models.load_model(model_name)
-                elif model_name.endswith('.h5'):
-                    MODEL = keras.models.load_model(model_name)
-                elif model_name.endswith('.json'):
-                    # Load model dari JSON + weights
-                    with open(model_name, 'r') as f:
-                        model_json = f.read()
-                    MODEL = keras.models.model_from_json(model_json)
-                    
-                    # Cari weights file
-                    weights_base = model_name.replace('.json', '')
-                    weights_files = [
-                        f"{weights_base}.h5",
-                        f"{weights_base}_weights.h5"
-                    ]
-                    for weights_file in weights_files:
-                        if os.path.exists(weights_file):
-                            MODEL.load_weights(weights_file)
-                            print(f"✓ Weights dimuat: {weights_file}")
-                            break
-                
-                print(f"✓ Model berhasil dimuat!")
-                print(f"  Model input shape: {MODEL.input_shape}")
-                print(f"  Number of layers: {len(MODEL.layers)}")
-                return True
-            except Exception as e:
-                print(f"✗ Error loading model: {str(e)}")
-                return False
-    
-    print("✗ Model tidak ditemukan!")
-    print("   Letakkan salah satu dari ini di folder yang sama dengan script ini:")
-    print("   - rice_leaf_disease_model.keras")
-    print("   - rice_leaf_disease_model.h5")
-    print("   - rice_leaf_disease_model.json (+ .h5 weights)")
-    return False
-
-
-def preprocess_image(image_data):
-    """
-    Preprocessing gambar dari base64 atau bytes
-    
-    Args:
-        image_data: base64 string atau bytes
-        
-    Returns:
-        Preprocessed image array atau None jika error
-    """
-    try:
-        # Jika string base64, decode dulu
-        if isinstance(image_data, str):
-            image_bytes = base64.b64decode(image_data)
-        else:
-            image_bytes = image_data
-
-        # Convert bytes ke image menggunakan PIL
-        image = Image.open(BytesIO(image_bytes))
-        
-        # Convert to RGB jika diperlukan
-        if image.mode != 'RGB':
-            image = image.convert('RGB')
-        
-        # Resize ke ukuran yang diharapkan model
-        image = image.resize(IMG_SIZE, Image.Resampling.LANCZOS)
-        
-        # Convert ke numpy array
-        img_array = np.array(image, dtype='float32')
-        
-        # Normalize pixel values ke range [0, 1]
-        img_array = img_array / 255.0
-        
-        # Add batch dimension
-        img_array = np.expand_dims(img_array, axis=0)
-        
-        return img_array
-        
-    except Exception as e:
-        print(f"Error preprocessing image: {str(e)}")
-        return None
-
-
-def classify_image(image_data):
-    """
-    Klasifikasi gambar menggunakan model
-    
-    Args:
-        image_data: base64 string atau bytes
-        
-    Returns:
-        Dict dengan hasil klasifikasi atau None jika error
-    """
-    # Mock mode - jika TensorFlow tidak tersedia
-    if not TENSORFLOW_AVAILABLE:
-        import random
-        predictions = [random.uniform(0.1, 0.9) for _ in CLASS_NAMES]
-        max_pred = max(predictions)
-        idx = predictions.index(max_pred)
-        
-        all_predictions = {}
-        for i, class_name in enumerate(CLASS_NAMES):
-            all_predictions[class_name] = round(predictions[i], 4)
-        
-        return {
-            'predicted_class': CLASS_NAMES[idx],
-            'confidence': round(max_pred, 4),
-            'all_predictions': all_predictions
-        }
-    
-    if MODEL is None:
-        return None
-    
-    try:
-        # Preprocess image
-        img_array = preprocess_image(image_data)
-        
-        if img_array is None:
-            return None
-        
-        # Prediction
-        predictions = MODEL.predict(img_array, verbose=0)
-        
-        # Get predicted class dan confidence
-        predicted_idx = np.argmax(predictions[0])
-        predicted_class = CLASS_NAMES[predicted_idx]
-        confidence = float(predictions[0][predicted_idx])
-        
-        # Build all predictions
-        all_predictions = {}
-        for idx, class_name in enumerate(CLASS_NAMES):
-            all_predictions[class_name] = float(predictions[0][idx])
-        
-        return {
-            'predicted_class': predicted_class,
-            'confidence': confidence,
-            'all_predictions': all_predictions
-        }
-        
-    except Exception as e:
-        print(f"Error during classification: {str(e)}")
-        return None
-
-
-# ============================================================================
-# ROUTES
-# ============================================================================
-
-@app.route('/health', methods=['POST', 'GET'])
-def health_check():
-    """Check apakah API berjalan dan model tersedia"""
-    if not TENSORFLOW_AVAILABLE:
-        return jsonify({
-            'status': 'ok',
-            'message': 'API running in MOCK MODE (TensorFlow not available)',
-            'model_loaded': False,
-            'mock_mode': True,
-            'classes': CLASS_NAMES
-        }), 200
-    
-    if MODEL is None:
-        return jsonify({
-            'status': 'error',
-            'message': 'Model not loaded',
-            'model_loaded': False
-        }), 503
-    
-    return jsonify({
-        'status': 'ok',
-        'message': 'API is running',
-        'model_loaded': True,
-        'model_path': MODEL_PATH,
-        'classes': CLASS_NAMES,
-        'input_shape': str(MODEL.input_shape)
-    }), 200
-
-
-@app.route('/classify', methods=['POST'])
-def classify():
-    """
-    API endpoint untuk klasifikasi gambar
-    
-    Expected request:
-    {
-        "image": "base64_encoded_image_string",
-        "filename": "optional_filename.jpg"
-    }
-    
-    Response:
-    {
-        "success": true,
-        "predicted_class": "Bacterialblight",
-        "confidence": 0.95,
-        "all_predictions": {
-            "Bacterialblight": 0.95,
-            "Brownspot": 0.04,
-            "Leafsmut": 0.01
-        }
-    }
-    """
-    try:
-        data = request.get_json()
-        
-        if data is None:
-            return jsonify({
-                'success': False,
-                'message': 'Request harus JSON'
-            }), 400
-        
-        # Validasi input
-        if 'image' not in data:
-            return jsonify({
-                'success': False,
-                'message': 'Field "image" (base64) diperlukan'
-            }), 400
-        
-        image_data = data['image']
-        filename = data.get('filename', 'unknown')
-        
-        # Klasifikasi
-        result = classify_image(image_data)
-        
-        if result is None:
-            return jsonify({
-                'success': False,
-                'message': 'Gagal memproses gambar'
-            }), 400
-        
-        # Log hasil
-        print(f"✓ Classification done: {filename} -> {result['predicted_class']} ({result['confidence']:.2%})")
-        
-        return jsonify({
-            'success': True,
-            'predicted_class': result['predicted_class'],
-            'confidence': result['confidence'],
-            'all_predictions': result['all_predictions'],
-            'filename': filename
-        }), 200
-        
-    except Exception as e:
-        print(f"✗ Error in classify endpoint: {str(e)}")
-        return jsonify({
-            'success': False,
-            'message': f'Error: {str(e)}'
-        }), 500
-
-
-@app.route('/classify-from-url', methods=['POST'])
-def classify_from_url():
-    """
-    Alternative endpoint untuk klasifikasi dari URL gambar
-    
-    Expected request:
-    {
-        "image_url": "http://example.com/image.jpg"
-    }
-    """
-    try:
-        data = request.get_json()
-        
-        if data is None or 'image_url' not in data:
-            return jsonify({
-                'success': False,
-                'message': 'Field "image_url" diperlukan'
-            }), 400
-        
-        image_url = data['image_url']
-        
-        import requests
-        response = requests.get(image_url, timeout=10)
-        
-        if response.status_code != 200:
-            return jsonify({
-                'success': False,
-                'message': f'Gagal download image dari URL'
-            }), 400
-        
-        # Klasifikasi
-        result = classify_image(response.content)
-        
-        if result is None:
-            return jsonify({
-                'success': False,
-                'message': 'Gagal memproses gambar'
-            }), 400
-        
-        return jsonify({
-            'success': True,
-            'predicted_class': result['predicted_class'],
-            'confidence': result['confidence'],
-            'all_predictions': result['all_predictions'],
-            'url': image_url
-        }), 200
-        
-    except Exception as e:
-        return jsonify({
-            'success': False,
-            'message': f'Error: {str(e)}'
-        }), 500
-
-
-@app.route('/info', methods=['GET'])
-def model_info():
-    """Get informasi tentang model"""
-    if MODEL is None:
-        return jsonify({
-            'status': 'error',
-            'message': 'Model not loaded'
-        }), 503
-    
-    return jsonify({
-        'model_loaded': True,
-        'model_path': MODEL_PATH,
-        'classes': CLASS_NAMES,
-        'number_of_classes': len(CLASS_NAMES),
-        'input_shape': str(MODEL.input_shape),
-        'number_of_layers': len(MODEL.layers),
-        'total_parameters': int(MODEL.count_params())
-    }), 200
-
-
-@app.route('/', methods=['GET'])
-def index():
-    """Root endpoint dengan informasi API"""
-    return jsonify({
-        'name': 'Rice Leaf Disease Classification API',
-        'version': '1.0',
-        'description': 'API untuk klasifikasi penyakit daun padi menggunakan CNN',
-        'endpoints': {
-            'POST /classify': 'Klasifikasi gambar (base64)',
-            'POST /classify-from-url': 'Klasifikasi gambar dari URL',
-            'GET /health': 'Health check',
-            'GET /info': 'Informasi model',
-            'GET /': 'Info API ini'
-        },
-        'model_status': 'Loaded' if MODEL is not None else 'Not loaded',
-        'classes': CLASS_NAMES
-    }), 200
-
-
-@app.errorhandler(404)
-def not_found(error):
-    return jsonify({
-        'success': False,
-        'message': 'Endpoint tidak ditemukan'
-    }), 404
-
-
-@app.errorhandler(500)
-def server_error(error):
-    return jsonify({
-        'success': False,
-        'message': 'Internal server error'
-    }), 500
-
-
-# ============================================================================
-# STARTUP
-# ============================================================================
-
-if __name__ == '__main__':
-    print("\n" + "="*60)
-    print("Rice Leaf Disease Classification API Server")
-    print("="*60 + "\n")
-    
-    # Pindah ke folder yang sama dengan script
-    script_dir = os.path.dirname(os.path.abspath(__file__))
-    os.chdir(script_dir)
-    print(f"Working directory: {os.getcwd()}\n")
-    
-    # Load model
-    print("Loading model...")
-    if not load_model():
-        print("\n⚠️  WARNING: Model tidak dapat dimuat!")
-        print("   API akan berjalan tapi endpoint /classify akan gagal.\n")
-    
-    print("\n" + "="*60)
-    print("Starting Flask API Server...")
-    print("="*60)
-    print("Server berjalan di http://127.0.0.1:5000/")
-    print("Tekan CTRL+C untuk menghentikan.\n")
-    
-    # Run Flask app
-    app.run(
-        host='127.0.0.1',
-        port=5000,
-        debug=False,  # Set ke True jika development
-        use_reloader=False
-    )
diff --git a/rice leaf diseases dataset/rice_leaf_cnn_classification.ipynb b/rice leaf diseases dataset/rice_leaf_cnn_classification.ipynb
index e3bcbe6..7ff5365 100644
--- a/rice leaf diseases dataset/rice_leaf_cnn_classification.ipynb	
+++ b/rice leaf diseases dataset/rice_leaf_cnn_classification.ipynb	
@@ -2,7 +2,7 @@
  "cells": [
   {
    "cell_type": "code",
-   "execution_count": 1,
+   "execution_count": 20,
    "id": "5ca735e6",
    "metadata": {},
    "outputs": [
@@ -10,7 +10,7 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "TensorFlow Version: 2.20.0\n",
+      "TensorFlow Version: 2.21.0\n",
       "GPU Available: []\n"
      ]
     }
@@ -69,7 +69,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 2,
+   "execution_count": 21,
    "id": "63695071",
    "metadata": {},
    "outputs": [
@@ -89,7 +89,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Loading Bacterialblight: 100%|█████████████████████████████████| 1604/1604 [00:02<00:00, 569.23it/s]\n"
+      "Loading Bacterialblight: 100%|█████████████████████████████████| 1604/1604 [00:03<00:00, 455.07it/s]\n"
      ]
     },
     {
@@ -106,28 +106,28 @@
      "text": [
       "\n",
       "Loading Brownspot:   0%|                                                   | 0/1620 [00:00<?, ?it/s]\u001b[A\n",
-      "Loading Brownspot:   4%|█▊                                       | 70/1620 [00:00<00:02, 699.62it/s]\u001b[A\n",
-      "Loading Brownspot:   9%|███▍                                    | 140/1620 [00:00<00:02, 699.30it/s]\u001b[A\n",
-      "Loading Brownspot:  13%|█████▏                                  | 211/1620 [00:00<00:02, 702.40it/s]\u001b[A\n",
-      "Loading Brownspot:  18%|███████▏                                | 289/1620 [00:00<00:01, 730.09it/s]\u001b[A\n",
-      "Loading Brownspot:  22%|████████▉                               | 363/1620 [00:00<00:01, 732.00it/s]\u001b[A\n",
-      "Loading Brownspot:  27%|██████████▊                             | 437/1620 [00:00<00:01, 713.49it/s]\u001b[A\n",
-      "Loading Brownspot:  31%|████████████▌                           | 509/1620 [00:00<00:01, 690.97it/s]\u001b[A\n",
-      "Loading Brownspot:  36%|██████████████▍                         | 584/1620 [00:00<00:01, 708.72it/s]\u001b[A\n",
-      "Loading Brownspot:  40%|████████████████▏                       | 656/1620 [00:00<00:01, 699.71it/s]\u001b[A\n",
-      "Loading Brownspot:  45%|█████████████████▉                      | 727/1620 [00:01<00:01, 690.09it/s]\u001b[A\n",
-      "Loading Brownspot:  49%|███████████████████▊                    | 801/1620 [00:01<00:01, 703.52it/s]\u001b[A\n",
-      "Loading Brownspot:  54%|█████████████████████▌                  | 872/1620 [00:01<00:01, 687.65it/s]\u001b[A\n",
-      "Loading Brownspot:  58%|███████████████████████▏                | 941/1620 [00:01<00:01, 669.68it/s]\u001b[A\n",
-      "Loading Brownspot:  62%|████████████████████████▎              | 1009/1620 [00:01<00:00, 660.11it/s]\u001b[A\n",
-      "Loading Brownspot:  66%|█████████████████████████▉             | 1076/1620 [00:01<00:00, 654.03it/s]\u001b[A\n",
-      "Loading Brownspot:  70%|███████████████████████████▍           | 1142/1620 [00:01<00:00, 634.46it/s]\u001b[A\n",
-      "Loading Brownspot:  75%|█████████████████████████████▎         | 1215/1620 [00:01<00:00, 659.71it/s]\u001b[A\n",
-      "Loading Brownspot:  79%|██████████████████████████████▊        | 1282/1620 [00:01<00:00, 642.89it/s]\u001b[A\n",
-      "Loading Brownspot:  83%|████████████████████████████████▍      | 1347/1620 [00:01<00:00, 625.76it/s]\u001b[A\n",
-      "Loading Brownspot:  88%|██████████████████████████████████▏    | 1418/1620 [00:02<00:00, 649.42it/s]\u001b[A\n",
-      "Loading Brownspot:  92%|███████████████████████████████████▊   | 1487/1620 [00:02<00:00, 660.73it/s]\u001b[A\n",
-      "Loading Brownspot: 100%|███████████████████████████████████████| 1620/1620 [00:02<00:00, 673.05it/s]\u001b[A\n"
+      "Loading Brownspot:   5%|█▉                                       | 75/1620 [00:00<00:02, 743.14it/s]\u001b[A\n",
+      "Loading Brownspot:   9%|███▋                                    | 150/1620 [00:00<00:02, 655.37it/s]\u001b[A\n",
+      "Loading Brownspot:  13%|█████▎                                  | 217/1620 [00:00<00:02, 612.44it/s]\u001b[A\n",
+      "Loading Brownspot:  18%|███████                                 | 284/1620 [00:00<00:02, 632.05it/s]\u001b[A\n",
+      "Loading Brownspot:  21%|████████▌                               | 348/1620 [00:00<00:02, 602.90it/s]\u001b[A\n",
+      "Loading Brownspot:  25%|██████████                              | 409/1620 [00:00<00:02, 584.49it/s]\u001b[A\n",
+      "Loading Brownspot:  29%|███████████▊                            | 477/1620 [00:00<00:01, 613.06it/s]\u001b[A\n",
+      "Loading Brownspot:  33%|█████████████▎                          | 539/1620 [00:00<00:01, 598.60it/s]\u001b[A\n",
+      "Loading Brownspot:  37%|██████████████▉                         | 606/1620 [00:00<00:01, 610.87it/s]\u001b[A\n",
+      "Loading Brownspot:  42%|████████████████▉                       | 684/1620 [00:01<00:01, 660.54it/s]\u001b[A\n",
+      "Loading Brownspot:  47%|██████████████████▋                     | 756/1620 [00:01<00:01, 674.30it/s]\u001b[A\n",
+      "Loading Brownspot:  51%|████████████████████▎                   | 825/1620 [00:01<00:01, 677.14it/s]\u001b[A\n",
+      "Loading Brownspot:  56%|██████████████████████▎                 | 904/1620 [00:01<00:01, 700.47it/s]\u001b[A\n",
+      "Loading Brownspot:  61%|████████████████████████▏               | 981/1620 [00:01<00:00, 716.94it/s]\u001b[A\n",
+      "Loading Brownspot:  65%|█████████████████████████▎             | 1053/1620 [00:01<00:00, 712.06it/s]\u001b[A\n",
+      "Loading Brownspot:  70%|███████████████████████████            | 1126/1620 [00:01<00:00, 711.62it/s]\u001b[A\n",
+      "Loading Brownspot:  74%|████████████████████████████▉          | 1204/1620 [00:01<00:00, 731.23it/s]\u001b[A\n",
+      "Loading Brownspot:  79%|██████████████████████████████▊        | 1278/1620 [00:01<00:00, 720.96it/s]\u001b[A\n",
+      "Loading Brownspot:  83%|████████████████████████████████▌      | 1351/1620 [00:02<00:00, 699.94it/s]\u001b[A\n",
+      "Loading Brownspot:  88%|██████████████████████████████████▎    | 1426/1620 [00:02<00:00, 706.58it/s]\u001b[A\n",
+      "Loading Brownspot:  93%|████████████████████████████████████▎  | 1508/1620 [00:02<00:00, 737.27it/s]\u001b[A\n",
+      "Loading Brownspot: 100%|███████████████████████████████████████| 1620/1620 [00:02<00:00, 683.08it/s]\u001b[A\n"
      ]
     },
     {
@@ -146,75 +146,51 @@
       "\n",
       "Loading Leafsmut:   0%|                                                    | 0/1460 [00:00<?, ?it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:   2%|▉                                         | 34/1460 [00:00<00:04, 336.45it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:   3%|█▎                                        | 46/1460 [00:00<00:03, 454.19it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:   6%|██▍                                       | 84/1460 [00:00<00:03, 432.14it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:   8%|███                                      | 110/1460 [00:00<00:02, 553.92it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:   9%|███▌                                     | 128/1460 [00:00<00:03, 370.47it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  12%|████▊                                    | 173/1460 [00:00<00:02, 588.14it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  12%|████▉                                    | 175/1460 [00:00<00:03, 405.81it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  16%|██████▌                                  | 232/1460 [00:00<00:02, 538.37it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  15%|██████                                   | 217/1460 [00:00<00:03, 378.27it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  20%|████████                                 | 287/1460 [00:00<00:02, 518.73it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  18%|███████▎                                 | 259/1460 [00:00<00:03, 389.27it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  24%|█████████▋                               | 346/1460 [00:00<00:02, 538.05it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  20%|████████▍                                | 299/1460 [00:00<00:03, 348.56it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  27%|███████████▎                             | 401/1460 [00:00<00:02, 525.77it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  24%|█████████▊                               | 349/1460 [00:00<00:02, 388.68it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  31%|████████████▋                            | 454/1460 [00:00<00:01, 524.78it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  27%|██████████▉                              | 390/1460 [00:01<00:02, 377.45it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  35%|██████████████▎                          | 510/1460 [00:00<00:01, 534.56it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  30%|████████████▏                            | 436/1460 [00:01<00:02, 398.68it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  39%|███████████████▊                         | 564/1460 [00:01<00:01, 535.09it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  33%|█████████████▍                           | 477/1460 [00:01<00:02, 383.61it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  43%|█████████████████▌                       | 624/1460 [00:01<00:01, 549.79it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  36%|██████████████▌                          | 520/1460 [00:01<00:02, 391.67it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  47%|███████████████████▍                     | 692/1460 [00:01<00:01, 587.94it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  38%|███████████████▋                         | 560/1460 [00:01<00:02, 383.14it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  52%|█████████████████████▎                   | 758/1460 [00:01<00:01, 608.61it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  41%|████████████████▉                        | 602/1460 [00:01<00:02, 388.36it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  57%|███████████████████████▏                 | 825/1460 [00:01<00:01, 625.59it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  44%|██████████████████                       | 642/1460 [00:01<00:02, 385.77it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  61%|████████████████████████▉                | 888/1460 [00:01<00:00, 614.82it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  47%|███████████████████▏                     | 682/1460 [00:01<00:02, 387.91it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  65%|██████████████████████████▋              | 950/1460 [00:01<00:00, 556.75it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  49%|████████████████████▏                    | 721/1460 [00:01<00:01, 382.54it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  69%|███████████████████████████▋            | 1011/1460 [00:01<00:00, 569.94it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  52%|█████████████████████▍                   | 764/1460 [00:01<00:01, 392.43it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  74%|█████████████████████████████▍          | 1076/1460 [00:01<00:00, 591.68it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  55%|██████████████████████▌                  | 804/1460 [00:02<00:01, 393.30it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  78%|███████████████████████████████▎        | 1144/1460 [00:02<00:00, 614.97it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  58%|███████████████████████▋                 | 844/1460 [00:02<00:01, 378.71it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  83%|█████████████████████████████████       | 1207/1460 [00:02<00:00, 610.18it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  60%|████████████████████████▊                | 883/1460 [00:02<00:01, 373.92it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  87%|██████████████████████████████████▊     | 1269/1460 [00:02<00:00, 598.20it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  63%|█████████████████████████▊               | 921/1460 [00:02<00:01, 375.59it/s]\u001b[A\u001b[A\n",
+      "Loading Leafsmut:  92%|████████████████████████████████████▌   | 1336/1460 [00:02<00:00, 614.32it/s]\u001b[A\u001b[A\n",
       "\n",
-      "Loading Leafsmut:  66%|██████████████████████████▉              | 959/1460 [00:02<00:01, 361.55it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  68%|████████████████████████████             | 998/1460 [00:02<00:01, 366.99it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  71%|████████████████████████████▎           | 1035/1460 [00:02<00:01, 351.99it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  74%|█████████████████████████████▌          | 1077/1460 [00:02<00:01, 369.48it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  76%|██████████████████████████████▌         | 1115/1460 [00:02<00:00, 352.80it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  79%|███████████████████████████████▊        | 1159/1460 [00:03<00:00, 377.04it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  82%|████████████████████████████████▊       | 1198/1460 [00:03<00:00, 366.30it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  85%|█████████████████████████████████▉      | 1238/1460 [00:03<00:00, 374.44it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  87%|██████████████████████████████████▉     | 1276/1460 [00:03<00:00, 360.79it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  90%|████████████████████████████████████    | 1318/1460 [00:03<00:00, 375.94it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  93%|█████████████████████████████████████▏  | 1356/1460 [00:03<00:00, 327.22it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut:  96%|██████████████████████████████████████▎ | 1398/1460 [00:03<00:00, 350.49it/s]\u001b[A\u001b[A\n",
-      "\n",
-      "Loading Leafsmut: 100%|████████████████████████████████████████| 1460/1460 [00:03<00:00, 375.37it/s]\u001b[A\u001b[A\n"
+      "Loading Leafsmut: 100%|████████████████████████████████████████| 1460/1460 [00:02<00:00, 582.97it/s]\u001b[A\u001b[A\n"
      ]
     },
     {
@@ -307,7 +283,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 22,
    "id": "a9a6d821",
    "metadata": {},
    "outputs": [
@@ -364,7 +340,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 23,
    "id": "f8de0b07",
    "metadata": {},
    "outputs": [
@@ -381,7 +357,7 @@
      "name": "stderr",
      "output_type": "stream",
      "text": [
-      "Resizing: 100%|███████████████████████████████████████████████| 4684/4684 [00:01<00:00, 3191.48it/s]\n"
+      "Resizing: 100%|███████████████████████████████████████████████| 4684/4684 [00:01<00:00, 3763.96it/s]\n"
      ]
     },
     {
@@ -444,7 +420,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 5,
+   "execution_count": 24,
    "id": "eba217e2",
    "metadata": {},
    "outputs": [
@@ -506,7 +482,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": 25,
    "id": "146d5d58",
    "metadata": {},
    "outputs": [
@@ -531,11 +507,11 @@
     {
      "data": {
       "text/html": [
-       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential\"</span>\n",
+       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"sequential_2\"</span>\n",
        "</pre>\n"
       ],
       "text/plain": [
-       "\u001b[1mModel: \"sequential\"\u001b[0m\n"
+       "\u001b[1mModel: \"sequential_2\"\u001b[0m\n"
       ]
      },
      "metadata": {},
@@ -547,80 +523,80 @@
        "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
        "┃<span style=\"font-weight: bold\"> Layer (type)                         </span>┃<span style=\"font-weight: bold\"> Output Shape                </span>┃<span style=\"font-weight: bold\">         Param # </span>┃\n",
        "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
-       "│ conv2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">896</span> │\n",
+       "│ conv2d_16 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">896</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │\n",
+       "│ batch_normalization_20               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │           <span style=\"color: #00af00; text-decoration-color: #00af00\">9,248</span> │\n",
+       "│ conv2d_17 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │           <span style=\"color: #00af00; text-decoration-color: #00af00\">9,248</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_1                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │\n",
+       "│ batch_normalization_21               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ max_pooling2d (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)         │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ max_pooling2d_8 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ dropout_12 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)        │          <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
+       "│ conv2d_18 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)        │          <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_2                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │\n",
+       "│ batch_normalization_22               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)        │          <span style=\"color: #00af00; text-decoration-color: #00af00\">36,928</span> │\n",
+       "│ conv2d_19 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)        │          <span style=\"color: #00af00; text-decoration-color: #00af00\">36,928</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_3                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │\n",
+       "│ batch_normalization_23               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">112</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ max_pooling2d_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ max_pooling2d_9 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ dropout_13 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │          <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │\n",
+       "│ conv2d_20 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │          <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_4                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │             <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │\n",
+       "│ batch_normalization_24               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │             <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │         <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │\n",
+       "│ conv2d_21 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │         <span style=\"color: #00af00; text-decoration-color: #00af00\">147,584</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_5                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │             <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │\n",
+       "│ batch_normalization_25               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">56</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │             <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ max_pooling2d_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ max_pooling2d_10 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ dropout_14 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │         <span style=\"color: #00af00; text-decoration-color: #00af00\">295,168</span> │\n",
+       "│ conv2d_22 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │         <span style=\"color: #00af00; text-decoration-color: #00af00\">295,168</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_6                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
+       "│ batch_normalization_26               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │         <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │\n",
+       "│ conv2d_23 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Conv2D</span>)                   │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │         <span style=\"color: #00af00; text-decoration-color: #00af00\">590,080</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_7                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
+       "│ batch_normalization_27               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">28</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ max_pooling2d_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)       │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ max_pooling2d_11 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">MaxPooling2D</span>)      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_3 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ dropout_15 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">14</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ global_average_pooling2d             │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ global_average_pooling2d_2           │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePooling2D</span>)             │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                        │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)                 │         <span style=\"color: #00af00; text-decoration-color: #00af00\">131,584</span> │\n",
+       "│ dense_6 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)                 │         <span style=\"color: #00af00; text-decoration-color: #00af00\">131,584</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_8                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)                 │           <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │\n",
+       "│ batch_normalization_28               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)                 │           <span style=\"color: #00af00; text-decoration-color: #00af00\">2,048</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ dropout_16 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">512</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dense_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)                 │         <span style=\"color: #00af00; text-decoration-color: #00af00\">131,328</span> │\n",
+       "│ dense_7 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)                 │         <span style=\"color: #00af00; text-decoration-color: #00af00\">131,328</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_9                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)                 │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
+       "│ batch_normalization_29               │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)                 │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,024</span> │\n",
        "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">BatchNormalization</span>)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_5 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                  │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
+       "│ dropout_17 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>)                 │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">256</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dense_2 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                   │             <span style=\"color: #00af00; text-decoration-color: #00af00\">771</span> │\n",
+       "│ dense_8 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>)                      │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)                   │             <span style=\"color: #00af00; text-decoration-color: #00af00\">771</span> │\n",
        "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
        "</pre>\n"
       ],
@@ -628,80 +604,80 @@
        "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\n",
        "┃\u001b[1m \u001b[0m\u001b[1mLayer (type)                        \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape               \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m        Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
        "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━┩\n",
-       "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │             \u001b[38;5;34m896\u001b[0m │\n",
+       "│ conv2d_16 (\u001b[38;5;33mConv2D\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │             \u001b[38;5;34m896\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │             \u001b[38;5;34m128\u001b[0m │\n",
+       "│ batch_normalization_20               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │             \u001b[38;5;34m128\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │           \u001b[38;5;34m9,248\u001b[0m │\n",
+       "│ conv2d_17 (\u001b[38;5;33mConv2D\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │           \u001b[38;5;34m9,248\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_1                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │             \u001b[38;5;34m128\u001b[0m │\n",
+       "│ batch_normalization_21               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │             \u001b[38;5;34m128\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ max_pooling2d_8 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout (\u001b[38;5;33mDropout\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ dropout_12 (\u001b[38;5;33mDropout\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_2 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │          \u001b[38;5;34m18,496\u001b[0m │\n",
+       "│ conv2d_18 (\u001b[38;5;33mConv2D\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │          \u001b[38;5;34m18,496\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_2                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │             \u001b[38;5;34m256\u001b[0m │\n",
+       "│ batch_normalization_22               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │             \u001b[38;5;34m256\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_3 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │          \u001b[38;5;34m36,928\u001b[0m │\n",
+       "│ conv2d_19 (\u001b[38;5;33mConv2D\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │          \u001b[38;5;34m36,928\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_3                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │             \u001b[38;5;34m256\u001b[0m │\n",
+       "│ batch_normalization_23               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m112\u001b[0m, \u001b[38;5;34m64\u001b[0m)        │             \u001b[38;5;34m256\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ max_pooling2d_9 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ dropout_13 (\u001b[38;5;33mDropout\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_4 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │          \u001b[38;5;34m73,856\u001b[0m │\n",
+       "│ conv2d_20 (\u001b[38;5;33mConv2D\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │          \u001b[38;5;34m73,856\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_4                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │             \u001b[38;5;34m512\u001b[0m │\n",
+       "│ batch_normalization_24               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │             \u001b[38;5;34m512\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_5 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │         \u001b[38;5;34m147,584\u001b[0m │\n",
+       "│ conv2d_21 (\u001b[38;5;33mConv2D\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │         \u001b[38;5;34m147,584\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_5                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │             \u001b[38;5;34m512\u001b[0m │\n",
+       "│ batch_normalization_25               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m56\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │             \u001b[38;5;34m512\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ max_pooling2d_2 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ max_pooling2d_10 (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_2 (\u001b[38;5;33mDropout\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ dropout_14 (\u001b[38;5;33mDropout\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_6 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │         \u001b[38;5;34m295,168\u001b[0m │\n",
+       "│ conv2d_22 (\u001b[38;5;33mConv2D\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │         \u001b[38;5;34m295,168\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_6                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │           \u001b[38;5;34m1,024\u001b[0m │\n",
+       "│ batch_normalization_26               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │           \u001b[38;5;34m1,024\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ conv2d_7 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │         \u001b[38;5;34m590,080\u001b[0m │\n",
+       "│ conv2d_23 (\u001b[38;5;33mConv2D\u001b[0m)                   │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │         \u001b[38;5;34m590,080\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_7                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │           \u001b[38;5;34m1,024\u001b[0m │\n",
+       "│ batch_normalization_27               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m28\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │           \u001b[38;5;34m1,024\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ max_pooling2d_3 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ max_pooling2d_11 (\u001b[38;5;33mMaxPooling2D\u001b[0m)      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_3 (\u001b[38;5;33mDropout\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ dropout_15 (\u001b[38;5;33mDropout\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m14\u001b[0m, \u001b[38;5;34m256\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ global_average_pooling2d             │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ global_average_pooling2d_2           │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
        "│ (\u001b[38;5;33mGlobalAveragePooling2D\u001b[0m)             │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dense (\u001b[38;5;33mDense\u001b[0m)                        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)                 │         \u001b[38;5;34m131,584\u001b[0m │\n",
+       "│ dense_6 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)                 │         \u001b[38;5;34m131,584\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_8                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)                 │           \u001b[38;5;34m2,048\u001b[0m │\n",
+       "│ batch_normalization_28               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)                 │           \u001b[38;5;34m2,048\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_4 (\u001b[38;5;33mDropout\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ dropout_16 (\u001b[38;5;33mDropout\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m512\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dense_1 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)                 │         \u001b[38;5;34m131,328\u001b[0m │\n",
+       "│ dense_7 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)                 │         \u001b[38;5;34m131,328\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ batch_normalization_9                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)                 │           \u001b[38;5;34m1,024\u001b[0m │\n",
+       "│ batch_normalization_29               │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)                 │           \u001b[38;5;34m1,024\u001b[0m │\n",
        "│ (\u001b[38;5;33mBatchNormalization\u001b[0m)                 │                             │                 │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dropout_5 (\u001b[38;5;33mDropout\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
+       "│ dropout_17 (\u001b[38;5;33mDropout\u001b[0m)                 │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m256\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
        "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
-       "│ dense_2 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m)                   │             \u001b[38;5;34m771\u001b[0m │\n",
+       "│ dense_8 (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m3\u001b[0m)                   │             \u001b[38;5;34m771\u001b[0m │\n",
        "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
       ]
      },
@@ -828,7 +804,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 28,
    "id": "240316b9",
    "metadata": {},
    "outputs": [
@@ -836,46 +812,44 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Compiling model...\n",
+      "Building MobileNetV2 Transfer Learning model...\n",
       "--------------------------------------------------\n",
-      "Configuration:\n",
-      "  - Optimizer: Adam (lr=0.001)\n",
-      "  - Loss: Categorical Crossentropy\n",
-      "  - Metrics: Accuracy\n",
-      "--------------------------------------------------\n",
-      "✓ Model compiled successfully!\n",
-      "\n"
+      "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/mobilenet_v2/mobilenet_v2_weights_tf_dim_ordering_tf_kernels_1.0_224_no_top.h5\n",
+      "\u001b[1m9406464/9406464\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 0us/step\n",
+      "✓ MobileNetV2 Model compiled successfully!\n"
      ]
     }
    ],
    "source": [
-    "# Option 2: Transfer Learning with MobileNetV2 (uncomment to use)\n",
-    "# base_model = MobileNetV2(input_shape=(IMG_SIZE, IMG_SIZE, 3), include_top=False, weights='imagenet')\n",
-    "# base_model.trainable = False\n",
-    "# \n",
-    "# model = models.Sequential([\n",
-    "#     base_model,\n",
-    "#     layers.GlobalAveragePooling2D(),\n",
-    "#     layers.Dense(256, activation='relu'),\n",
-    "#     layers.Dropout(0.5),\n",
-    "#     layers.Dense(len(classes), activation='softmax')\n",
-    "# ])\n",
-    "\n",
-    "# Compile the model\n",
-    "print(\"Compiling model...\")\n",
+    "# MENGGUNAKAN TRANSFER LEARNING (MOBILENETV2)\n",
+    "print(\"Building MobileNetV2 Transfer Learning model...\")\n",
     "print(\"-\" * 50)\n",
+    "\n",
+    "# 1. Ambil model yang sudah pintar (Pre-trained)\n",
+    "base_model = MobileNetV2(input_shape=(IMG_SIZE, IMG_SIZE, 3), \n",
+    "                         include_top=False, \n",
+    "                         weights='imagenet')\n",
+    "\n",
+    "# 2. Bekukan model dasar agar tidak berubah saat training awal\n",
+    "base_model.trainable = False \n",
+    "\n",
+    "# 3. Rakit arsitektur model baru\n",
+    "model = models.Sequential([\n",
+    "    base_model,\n",
+    "    layers.GlobalAveragePooling2D(),\n",
+    "    layers.Dense(128, activation='relu'),\n",
+    "    layers.Dropout(0.3), # Mencegah overfitting (penting untuk TA!)\n",
+    "    layers.Dense(len(classes), activation='softmax')\n",
+    "])\n",
+    "\n",
+    "# 4. Compile model dengan Learning Rate yang lebih kecil (0.0001) agar stabil\n",
     "model.compile(\n",
-    "    optimizer=keras.optimizers.Adam(learning_rate=0.001),\n",
+    "    optimizer=keras.optimizers.Adam(learning_rate=0.0001), \n",
     "    loss='categorical_crossentropy',\n",
     "    metrics=['accuracy']\n",
     ")\n",
     "\n",
-    "print(\"Configuration:\")\n",
-    "print(\"  - Optimizer: Adam (lr=0.001)\")\n",
-    "print(\"  - Loss: Categorical Crossentropy\")\n",
-    "print(\"  - Metrics: Accuracy\")\n",
-    "print(\"-\" * 50)\n",
-    "print(\"✓ Model compiled successfully!\\n\")"
+    "print(\"✓ MobileNetV2 Model compiled successfully!\")"
    ]
   },
   {
@@ -888,7 +862,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 27,
    "id": "26699eff",
    "metadata": {},
    "outputs": [
@@ -907,60 +881,51 @@
       "\n",
       "📊 Training Progress:\n",
       "--------------------------------------------------------------------------------\n",
-      "[█░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 1/50 | Loss: 0.9152 | Val Loss: 2.1880 | Acc: 0.6974 | Val Acc: 0.3115 | ETA: 15860s\n",
-      "[██░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 2/50 | Loss: 0.5917 | Val Loss: 5.0228 | Acc: 0.7959 | Val Acc: 0.3457 | ETA: 14999s\n",
-      "[███░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 3/50 | Loss: 0.4868 | Val Loss: 3.4162 | Acc: 0.8270 | Val Acc: 0.5164 | ETA: 14541s\n",
-      "[████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 4/50 | Loss: 0.3732 | Val Loss: 1.2514 | Acc: 0.8630 | Val Acc: 0.6159 | ETA: 14109s\n",
-      "[█████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 5/50 | Loss: 0.3450 | Val Loss: 1.0146 | Acc: 0.8661 | Val Acc: 0.7454 | ETA: 13739s\n",
-      "[██████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 6/50 | Loss: 0.2968 | Val Loss: 0.5368 | Acc: 0.8877 | Val Acc: 0.7923 | ETA: 13377s\n",
-      "[███████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 7/50 | Loss: 0.2895 | Val Loss: 0.7549 | Acc: 0.8990 | Val Acc: 0.7767 | ETA: 13036s\n",
-      "[████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 8/50 | Loss: 0.2722 | Val Loss: 0.4136 | Acc: 0.9030 | Val Acc: 0.8876 | ETA: 12708s\n",
-      "[█████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 9/50 | Loss: 0.2799 | Val Loss: 1.5129 | Acc: 0.9015 | Val Acc: 0.6743 | ETA: 12383s\n",
-      "[██████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 10/50 | Loss: 0.2388 | Val Loss: 0.7123 | Acc: 0.9152 | Val Acc: 0.7653 | ETA: 12061s\n",
-      "[███████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 11/50 | Loss: 0.1875 | Val Loss: 0.1276 | Acc: 0.9317 | Val Acc: 0.9545 | ETA: 11745s\n",
-      "[████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 12/50 | Loss: 0.2415 | Val Loss: 0.3129 | Acc: 0.9146 | Val Acc: 0.8834 | ETA: 11439s\n",
-      "[█████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 13/50 | Loss: 0.1797 | Val Loss: 0.2248 | Acc: 0.9347 | Val Acc: 0.9175 | ETA: 11128s\n",
-      "[██████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 14/50 | Loss: 0.1990 | Val Loss: 0.0882 | Acc: 0.9295 | Val Acc: 0.9758 | ETA: 10823s\n",
-      "[███████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 15/50 | Loss: 0.1830 | Val Loss: 0.3407 | Acc: 0.9359 | Val Acc: 0.8933 | ETA: 10523s\n",
-      "[████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 16/50 | Loss: 0.1374 | Val Loss: 0.0735 | Acc: 0.9524 | Val Acc: 0.9673 | ETA: 10227s\n",
-      "[█████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 17/50 | Loss: 0.1476 | Val Loss: 0.3952 | Acc: 0.9491 | Val Acc: 0.8905 | ETA: 9939s\n",
-      "[██████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 18/50 | Loss: 0.1439 | Val Loss: 0.1214 | Acc: 0.9484 | Val Acc: 0.9659 | ETA: 9689s\n",
-      "[███████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 19/50 | Loss: 0.1787 | Val Loss: 7.9704 | Acc: 0.9417 | Val Acc: 0.4523 | ETA: 9383s\n",
-      "[████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 20/50 | Loss: 0.1424 | Val Loss: 0.1060 | Acc: 0.9518 | Val Acc: 0.9701 | ETA: 9076s\n",
-      "[█████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 21/50 | Loss: 0.1164 | Val Loss: 1.1432 | Acc: 0.9585 | Val Acc: 0.7468 | ETA: 8772s\n",
-      "[██████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 22/50 | Loss: 0.1030 | Val Loss: 0.0563 | Acc: 0.9634 | Val Acc: 0.9844 | ETA: 8473s\n",
-      "[███████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 23/50 | Loss: 0.0679 | Val Loss: 0.0226 | Acc: 0.9771 | Val Acc: 0.9957 | ETA: 8181s\n",
-      "[████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 24/50 | Loss: 0.0753 | Val Loss: 0.0315 | Acc: 0.9744 | Val Acc: 0.9915 | ETA: 7888s\n",
-      "[█████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 25/50 | Loss: 0.0921 | Val Loss: 0.1533 | Acc: 0.9671 | Val Acc: 0.9772 | ETA: 7606s\n",
-      "[██████████████████████████░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 26/50 | Loss: 0.0928 | Val Loss: 0.1205 | Acc: 0.9725 | Val Acc: 0.9673 | ETA: 7318s\n",
-      "[███████████████████████████░░░░░░░░░░░░░░░░░░░░░░░] Epoch 27/50 | Loss: 0.0654 | Val Loss: 0.0055 | Acc: 0.9777 | Val Acc: 1.0000 | ETA: 7022s\n",
-      "[████████████████████████████░░░░░░░░░░░░░░░░░░░░░░] Epoch 28/50 | Loss: 0.0633 | Val Loss: 0.0072 | Acc: 0.9780 | Val Acc: 0.9986 | ETA: 6719s\n",
-      "[█████████████████████████████░░░░░░░░░░░░░░░░░░░░░] Epoch 29/50 | Loss: 0.0886 | Val Loss: 0.0197 | Acc: 0.9728 | Val Acc: 0.9943 | ETA: 6414s\n",
-      "[██████████████████████████████░░░░░░░░░░░░░░░░░░░░] Epoch 30/50 | Loss: 0.0609 | Val Loss: 0.0272 | Acc: 0.9777 | Val Acc: 0.9915 | ETA: 6111s\n",
-      "[███████████████████████████████░░░░░░░░░░░░░░░░░░░] Epoch 31/50 | Loss: 0.0593 | Val Loss: 0.0490 | Acc: 0.9783 | Val Acc: 0.9730 | ETA: 5806s\n",
-      "[████████████████████████████████░░░░░░░░░░░░░░░░░░] Epoch 32/50 | Loss: 0.0648 | Val Loss: 0.0576 | Acc: 0.9808 | Val Acc: 0.9716 | ETA: 5498s\n",
-      "[█████████████████████████████████░░░░░░░░░░░░░░░░░] Epoch 33/50 | Loss: 0.0487 | Val Loss: 0.0035 | Acc: 0.9823 | Val Acc: 1.0000 | ETA: 5190s\n",
-      "[██████████████████████████████████░░░░░░░░░░░░░░░░] Epoch 34/50 | Loss: 0.0389 | Val Loss: 0.0049 | Acc: 0.9872 | Val Acc: 0.9986 | ETA: 4883s\n",
-      "[███████████████████████████████████░░░░░░░░░░░░░░░] Epoch 35/50 | Loss: 0.0398 | Val Loss: 0.0015 | Acc: 0.9887 | Val Acc: 1.0000 | ETA: 4575s\n",
-      "[████████████████████████████████████░░░░░░░░░░░░░░] Epoch 36/50 | Loss: 0.0331 | Val Loss: 0.0051 | Acc: 0.9893 | Val Acc: 0.9986 | ETA: 4269s\n",
-      "[█████████████████████████████████████░░░░░░░░░░░░░] Epoch 37/50 | Loss: 0.0446 | Val Loss: 0.0020 | Acc: 0.9832 | Val Acc: 1.0000 | ETA: 3962s\n",
-      "[██████████████████████████████████████░░░░░░░░░░░░] Epoch 38/50 | Loss: 0.0367 | Val Loss: 0.0035 | Acc: 0.9863 | Val Acc: 1.0000 | ETA: 3657s\n",
-      "[███████████████████████████████████████░░░░░░░░░░░] Epoch 39/50 | Loss: 0.0369 | Val Loss: 0.0012 | Acc: 0.9875 | Val Acc: 1.0000 | ETA: 3351s\n",
-      "[████████████████████████████████████████░░░░░░░░░░] Epoch 40/50 | Loss: 0.0287 | Val Loss: 0.0213 | Acc: 0.9912 | Val Acc: 0.9943 | ETA: 3045s\n",
-      "[█████████████████████████████████████████░░░░░░░░░] Epoch 41/50 | Loss: 0.0434 | Val Loss: 0.0022 | Acc: 0.9851 | Val Acc: 1.0000 | ETA: 2739s\n",
-      "[██████████████████████████████████████████░░░░░░░░] Epoch 42/50 | Loss: 0.0456 | Val Loss: 0.1725 | Acc: 0.9841 | Val Acc: 0.9459 | ETA: 2433s\n",
-      "[███████████████████████████████████████████░░░░░░░] Epoch 43/50 | Loss: 0.0413 | Val Loss: 0.0035 | Acc: 0.9863 | Val Acc: 1.0000 | ETA: 2128s\n",
-      "[████████████████████████████████████████████░░░░░░] Epoch 44/50 | Loss: 0.0283 | Val Loss: 0.0036 | Acc: 0.9908 | Val Acc: 1.0000 | ETA: 1826s\n",
-      "[█████████████████████████████████████████████░░░░░] Epoch 45/50 | Loss: 0.0326 | Val Loss: 0.0004 | Acc: 0.9893 | Val Acc: 1.0000 | ETA: 1521s\n",
-      "[██████████████████████████████████████████████░░░░] Epoch 46/50 | Loss: 0.0203 | Val Loss: 0.0003 | Acc: 0.9921 | Val Acc: 1.0000 | ETA: 1216s\n",
-      "[███████████████████████████████████████████████░░░] Epoch 47/50 | Loss: 0.0204 | Val Loss: 0.0010 | Acc: 0.9942 | Val Acc: 1.0000 | ETA: 912s\n",
-      "[████████████████████████████████████████████████░░] Epoch 48/50 | Loss: 0.0355 | Val Loss: 0.0006 | Acc: 0.9863 | Val Acc: 1.0000 | ETA: 607s\n",
-      "[█████████████████████████████████████████████████░] Epoch 49/50 | Loss: 0.0400 | Val Loss: 0.0006 | Acc: 0.9887 | Val Acc: 1.0000 | ETA: 303s\n",
-      "[██████████████████████████████████████████████████] Epoch 50/50 | Loss: 0.0168 | Val Loss: 0.0004 | Acc: 0.9939 | Val Acc: 1.0000 | ETA: 0s\n",
-      "--------------------------------------------------------------------------------\n",
-      "✓ Training completed in 15182s\n",
-      "\n",
-      "✅ Model training and validation completed!\n"
+      "[█░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 1/50 | Loss: 0.8581 | Val Loss: 2.1125 | Acc: 0.7004 | Val Acc: 0.3457 | ETA: 15230s\n",
+      "[██░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 2/50 | Loss: 0.5727 | Val Loss: 4.5162 | Acc: 0.7984 | Val Acc: 0.4538 | ETA: 14684s\n",
+      "[███░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 3/50 | Loss: 0.4536 | Val Loss: 4.7104 | Acc: 0.8331 | Val Acc: 0.2489 | ETA: 14318s\n",
+      "[████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 4/50 | Loss: 0.3599 | Val Loss: 1.9790 | Acc: 0.8618 | Val Acc: 0.5818 | ETA: 13955s\n",
+      "[█████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 5/50 | Loss: 0.3339 | Val Loss: 1.0550 | Acc: 0.8746 | Val Acc: 0.6828 | ETA: 13671s\n",
+      "[██████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 6/50 | Loss: 0.3125 | Val Loss: 0.3040 | Acc: 0.8841 | Val Acc: 0.8890 | ETA: 13361s\n",
+      "[███████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 7/50 | Loss: 0.2461 | Val Loss: 0.4891 | Acc: 0.9097 | Val Acc: 0.8478 | ETA: 13062s\n",
+      "[████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 8/50 | Loss: 0.2662 | Val Loss: 0.2847 | Acc: 0.9024 | Val Acc: 0.8962 | ETA: 12766s\n",
+      "[█████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 9/50 | Loss: 0.1980 | Val Loss: 0.3768 | Acc: 0.9311 | Val Acc: 0.8791 | ETA: 12465s\n",
+      "[██████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 10/50 | Loss: 0.2186 | Val Loss: 0.5156 | Acc: 0.9185 | Val Acc: 0.8080 | ETA: 12159s\n",
+      "[███████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 11/50 | Loss: 0.2260 | Val Loss: 0.4154 | Acc: 0.9234 | Val Acc: 0.8720 | ETA: 11857s\n",
+      "[████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 12/50 | Loss: 0.1918 | Val Loss: 1.4041 | Acc: 0.9335 | Val Acc: 0.7013 | ETA: 11566s\n",
+      "[█████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 13/50 | Loss: 0.1812 | Val Loss: 0.9136 | Acc: 0.9320 | Val Acc: 0.7169 | ETA: 11274s\n",
+      "[██████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 14/50 | Loss: 0.1381 | Val Loss: 0.0307 | Acc: 0.9515 | Val Acc: 0.9972 | ETA: 10974s\n",
+      "[███████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 15/50 | Loss: 0.1227 | Val Loss: 0.0818 | Acc: 0.9561 | Val Acc: 0.9716 | ETA: 10688s\n",
+      "[████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 16/50 | Loss: 0.1217 | Val Loss: 0.0901 | Acc: 0.9542 | Val Acc: 0.9744 | ETA: 10397s\n",
+      "[█████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 17/50 | Loss: 0.1140 | Val Loss: 0.1562 | Acc: 0.9588 | Val Acc: 0.9331 | ETA: 10112s\n",
+      "[██████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 18/50 | Loss: 0.0963 | Val Loss: 0.1386 | Acc: 0.9664 | Val Acc: 0.9531 | ETA: 9808s\n",
+      "[███████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 19/50 | Loss: 0.1015 | Val Loss: 0.0694 | Acc: 0.9655 | Val Acc: 0.9659 | ETA: 9503s\n",
+      "[████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 20/50 | Loss: 0.0768 | Val Loss: 0.0260 | Acc: 0.9722 | Val Acc: 0.9929 | ETA: 9198s\n",
+      "[█████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 21/50 | Loss: 0.0736 | Val Loss: 0.0410 | Acc: 0.9747 | Val Acc: 0.9829 | ETA: 8893s\n",
+      "[██████████████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░] Epoch 22/50 | Loss: 0.0580 | Val Loss: 0.0110 | Acc: 0.9805 | Val Acc: 0.9986 | ETA: 8584s\n"
+     ]
+    },
+    {
+     "ename": "KeyboardInterrupt",
+     "evalue": "",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[31m---------------------------------------------------------------------------\u001b[39m",
+      "\u001b[31mKeyboardInterrupt\u001b[39m                         Traceback (most recent call last)",
+      "\u001b[36mCell\u001b[39m\u001b[36m \u001b[39m\u001b[32mIn[27]\u001b[39m\u001b[32m, line 65\u001b[39m\n\u001b[32m     62\u001b[39m \u001b[38;5;28mprint\u001b[39m(\u001b[33mf\u001b[39m\u001b[33m\"\u001b[39m\u001b[33mLoss Function: Categorical Crossentropy\u001b[39m\u001b[33m\"\u001b[39m)\n\u001b[32m     63\u001b[39m \u001b[38;5;28mprint\u001b[39m(\u001b[33m\"\u001b[39m\u001b[33m=\u001b[39m\u001b[33m\"\u001b[39m * \u001b[32m80\u001b[39m)\n\u001b[32m---> \u001b[39m\u001b[32m65\u001b[39m history = \u001b[43mmodel\u001b[49m\u001b[43m.\u001b[49m\u001b[43mfit\u001b[49m\u001b[43m(\u001b[49m\n\u001b[32m     66\u001b[39m \u001b[43m    \u001b[49m\u001b[43mtrain_datagen\u001b[49m\u001b[43m.\u001b[49m\u001b[43mflow\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX_train\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43my_train_cat\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[43m=\u001b[49m\u001b[43mBATCH_SIZE\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m     67\u001b[39m \u001b[43m    \u001b[49m\u001b[43mepochs\u001b[49m\u001b[43m=\u001b[49m\u001b[43mEPOCHS\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m     68\u001b[39m \u001b[43m    \u001b[49m\u001b[43mbatch_size\u001b[49m\u001b[43m=\u001b[49m\u001b[43mBATCH_SIZE\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m     69\u001b[39m \u001b[43m    \u001b[49m\u001b[43mvalidation_data\u001b[49m\u001b[43m=\u001b[49m\u001b[43m(\u001b[49m\u001b[43mX_val\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43my_val_cat\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m     70\u001b[39m \u001b[43m    \u001b[49m\u001b[43mcallbacks\u001b[49m\u001b[43m=\u001b[49m\u001b[43m[\u001b[49m\u001b[43mearly_stop\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mreduce_lr\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mProgressCallback\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m]\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m     71\u001b[39m \u001b[43m    \u001b[49m\u001b[43mverbose\u001b[49m\u001b[43m=\u001b[49m\u001b[32;43m0\u001b[39;49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# Suppress default verbose output\u001b[39;49;00m\n\u001b[32m     72\u001b[39m \u001b[43m)\u001b[49m\n\u001b[32m     74\u001b[39m \u001b[38;5;28mprint\u001b[39m(\u001b[33m\"\u001b[39m\u001b[33m✅ Model training and validation completed!\u001b[39m\u001b[33m\"\u001b[39m)\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\keras\\src\\utils\\traceback_utils.py:117\u001b[39m, in \u001b[36mfilter_traceback.<locals>.error_handler\u001b[39m\u001b[34m(*args, **kwargs)\u001b[39m\n\u001b[32m    115\u001b[39m filtered_tb = \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[32m    116\u001b[39m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[32m--> \u001b[39m\u001b[32m117\u001b[39m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[43m*\u001b[49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m*\u001b[49m\u001b[43m*\u001b[49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m    118\u001b[39m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[32m    119\u001b[39m     filtered_tb = _process_traceback_frames(e.__traceback__)\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\keras\\src\\backend\\tensorflow\\trainer.py:399\u001b[39m, in \u001b[36mTensorFlowTrainer.fit\u001b[39m\u001b[34m(self, x, y, batch_size, epochs, verbose, callbacks, validation_split, validation_data, shuffle, class_weight, sample_weight, initial_epoch, steps_per_epoch, validation_steps, validation_batch_size, validation_freq)\u001b[39m\n\u001b[32m    397\u001b[39m \u001b[38;5;28;01mfor\u001b[39;00m begin_step, end_step, iterator \u001b[38;5;129;01min\u001b[39;00m epoch_iterator:\n\u001b[32m    398\u001b[39m     callbacks.on_train_batch_begin(begin_step)\n\u001b[32m--> \u001b[39m\u001b[32m399\u001b[39m     logs = \u001b[38;5;28;43mself\u001b[39;49m\u001b[43m.\u001b[49m\u001b[43mtrain_function\u001b[49m\u001b[43m(\u001b[49m\u001b[43miterator\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m    400\u001b[39m     callbacks.on_train_batch_end(end_step, logs)\n\u001b[32m    401\u001b[39m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m.stop_training:\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\keras\\src\\backend\\tensorflow\\trainer.py:241\u001b[39m, in \u001b[36mTensorFlowTrainer._make_function.<locals>.function\u001b[39m\u001b[34m(iterator)\u001b[39m\n\u001b[32m    237\u001b[39m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34mfunction\u001b[39m(iterator):\n\u001b[32m    238\u001b[39m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(\n\u001b[32m    239\u001b[39m         iterator, (tf.data.Iterator, tf.distribute.DistributedIterator)\n\u001b[32m    240\u001b[39m     ):\n\u001b[32m--> \u001b[39m\u001b[32m241\u001b[39m         opt_outputs = \u001b[43mmulti_step_on_iterator\u001b[49m\u001b[43m(\u001b[49m\u001b[43miterator\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m    242\u001b[39m         \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m opt_outputs.has_value():\n\u001b[32m    243\u001b[39m             \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mStopIteration\u001b[39;00m\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\tensorflow\\python\\util\\traceback_utils.py:164\u001b[39m, in \u001b[36mfilter_traceback.<locals>.error_handler\u001b[39m\u001b[34m(*args, **kwargs)\u001b[39m\n\u001b[32m    162\u001b[39m filtered_tb = \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[32m    163\u001b[39m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[32m--> \u001b[39m\u001b[32m164\u001b[39m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfn\u001b[49m\u001b[43m(\u001b[49m\u001b[43m*\u001b[49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m*\u001b[49m\u001b[43m*\u001b[49m\u001b[43mkwargs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m    165\u001b[39m \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mException\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[32m    166\u001b[39m   filtered_tb = _process_traceback_frames(e.__traceback__)\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\polymorphic_function.py:833\u001b[39m, in \u001b[36mFunction.__call__\u001b[39m\u001b[34m(self, *args, **kwds)\u001b[39m\n\u001b[32m    830\u001b[39m compiler = \u001b[33m\"\u001b[39m\u001b[33mxla\u001b[39m\u001b[33m\"\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m._jit_compile \u001b[38;5;28;01melse\u001b[39;00m \u001b[33m\"\u001b[39m\u001b[33mnonXla\u001b[39m\u001b[33m\"\u001b[39m\n\u001b[32m    832\u001b[39m \u001b[38;5;28;01mwith\u001b[39;00m OptionalXlaContext(\u001b[38;5;28mself\u001b[39m._jit_compile):\n\u001b[32m--> \u001b[39m\u001b[32m833\u001b[39m   result = \u001b[38;5;28;43mself\u001b[39;49m\u001b[43m.\u001b[49m\u001b[43m_call\u001b[49m\u001b[43m(\u001b[49m\u001b[43m*\u001b[49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43m*\u001b[49m\u001b[43m*\u001b[49m\u001b[43mkwds\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m    835\u001b[39m new_tracing_count = \u001b[38;5;28mself\u001b[39m.experimental_get_tracing_count()\n\u001b[32m    836\u001b[39m without_tracing = (tracing_count == new_tracing_count)\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\polymorphic_function.py:878\u001b[39m, in \u001b[36mFunction._call\u001b[39m\u001b[34m(self, *args, **kwds)\u001b[39m\n\u001b[32m    875\u001b[39m \u001b[38;5;28mself\u001b[39m._lock.release()\n\u001b[32m    876\u001b[39m \u001b[38;5;66;03m# In this case we have not created variables on the first call. So we can\u001b[39;00m\n\u001b[32m    877\u001b[39m \u001b[38;5;66;03m# run the first trace but we should fail if variables are created.\u001b[39;00m\n\u001b[32m--> \u001b[39m\u001b[32m878\u001b[39m results = \u001b[43mtracing_compilation\u001b[49m\u001b[43m.\u001b[49m\u001b[43mcall_function\u001b[49m\u001b[43m(\u001b[49m\n\u001b[32m    879\u001b[39m \u001b[43m    \u001b[49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mkwds\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m.\u001b[49m\u001b[43m_variable_creation_config\u001b[49m\n\u001b[32m    880\u001b[39m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m    881\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m._created_variables:\n\u001b[32m    882\u001b[39m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\u001b[33m\"\u001b[39m\u001b[33mCreating variables on a non-first call to a function\u001b[39m\u001b[33m\"\u001b[39m\n\u001b[32m    883\u001b[39m                    \u001b[33m\"\u001b[39m\u001b[33m decorated with tf.function.\u001b[39m\u001b[33m\"\u001b[39m)\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\tracing_compilation.py:139\u001b[39m, in \u001b[36mcall_function\u001b[39m\u001b[34m(args, kwargs, tracing_options)\u001b[39m\n\u001b[32m    137\u001b[39m bound_args = function.function_type.bind(*args, **kwargs)\n\u001b[32m    138\u001b[39m flat_inputs = function.function_type.unpack_inputs(bound_args)\n\u001b[32m--> \u001b[39m\u001b[32m139\u001b[39m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mfunction\u001b[49m\u001b[43m.\u001b[49m\u001b[43m_call_flat\u001b[49m\u001b[43m(\u001b[49m\u001b[43m  \u001b[49m\u001b[38;5;66;43;03m# pylint: disable=protected-access\u001b[39;49;00m\n\u001b[32m    140\u001b[39m \u001b[43m    \u001b[49m\u001b[43mflat_inputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcaptured_inputs\u001b[49m\u001b[43m=\u001b[49m\u001b[43mfunction\u001b[49m\u001b[43m.\u001b[49m\u001b[43mcaptured_inputs\u001b[49m\n\u001b[32m    141\u001b[39m \u001b[43m\u001b[49m\u001b[43m)\u001b[49m\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\concrete_function.py:1322\u001b[39m, in \u001b[36mConcreteFunction._call_flat\u001b[39m\u001b[34m(self, tensor_inputs, captured_inputs)\u001b[39m\n\u001b[32m   1318\u001b[39m possible_gradient_type = gradients_util.PossibleTapeGradientTypes(args)\n\u001b[32m   1319\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m (possible_gradient_type == gradients_util.POSSIBLE_GRADIENT_TYPES_NONE\n\u001b[32m   1320\u001b[39m     \u001b[38;5;129;01mand\u001b[39;00m executing_eagerly):\n\u001b[32m   1321\u001b[39m   \u001b[38;5;66;03m# No tape is watching; skip to running the function.\u001b[39;00m\n\u001b[32m-> \u001b[39m\u001b[32m1322\u001b[39m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;43mself\u001b[39;49m\u001b[43m.\u001b[49m\u001b[43m_inference_function\u001b[49m\u001b[43m.\u001b[49m\u001b[43mcall_preflattened\u001b[49m\u001b[43m(\u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m   1323\u001b[39m forward_backward = \u001b[38;5;28mself\u001b[39m._select_forward_and_backward_functions(\n\u001b[32m   1324\u001b[39m     args,\n\u001b[32m   1325\u001b[39m     possible_gradient_type,\n\u001b[32m   1326\u001b[39m     executing_eagerly)\n\u001b[32m   1327\u001b[39m forward_function, args_with_tangents = forward_backward.forward()\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\atomic_function.py:216\u001b[39m, in \u001b[36mAtomicFunction.call_preflattened\u001b[39m\u001b[34m(self, args)\u001b[39m\n\u001b[32m    214\u001b[39m \u001b[38;5;28;01mdef\u001b[39;00m\u001b[38;5;250m \u001b[39m\u001b[34mcall_preflattened\u001b[39m(\u001b[38;5;28mself\u001b[39m, args: Sequence[core.Tensor]) -> Any:\n\u001b[32m    215\u001b[39m \u001b[38;5;250m  \u001b[39m\u001b[33;03m\"\"\"Calls with flattened tensor inputs and returns the structured output.\"\"\"\u001b[39;00m\n\u001b[32m--> \u001b[39m\u001b[32m216\u001b[39m   flat_outputs = \u001b[38;5;28;43mself\u001b[39;49m\u001b[43m.\u001b[49m\u001b[43mcall_flat\u001b[49m\u001b[43m(\u001b[49m\u001b[43m*\u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m    217\u001b[39m   \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28mself\u001b[39m.function_type.pack_output(flat_outputs)\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\polymorphic_function\\atomic_function.py:251\u001b[39m, in \u001b[36mAtomicFunction.call_flat\u001b[39m\u001b[34m(self, *args)\u001b[39m\n\u001b[32m    249\u001b[39m \u001b[38;5;28;01mwith\u001b[39;00m record.stop_recording():\n\u001b[32m    250\u001b[39m   \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mself\u001b[39m._bound_context.executing_eagerly():\n\u001b[32m--> \u001b[39m\u001b[32m251\u001b[39m     outputs = \u001b[38;5;28;43mself\u001b[39;49m\u001b[43m.\u001b[49m\u001b[43m_bound_context\u001b[49m\u001b[43m.\u001b[49m\u001b[43mcall_function\u001b[49m\u001b[43m(\u001b[49m\n\u001b[32m    252\u001b[39m \u001b[43m        \u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m.\u001b[49m\u001b[43mname\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m    253\u001b[39m \u001b[43m        \u001b[49m\u001b[38;5;28;43mlist\u001b[39;49m\u001b[43m(\u001b[49m\u001b[43margs\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m    254\u001b[39m \u001b[43m        \u001b[49m\u001b[38;5;28;43mlen\u001b[39;49m\u001b[43m(\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m.\u001b[49m\u001b[43mfunction_type\u001b[49m\u001b[43m.\u001b[49m\u001b[43mflat_outputs\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m    255\u001b[39m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m    256\u001b[39m   \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[32m    257\u001b[39m     outputs = make_call_op_in_graph(\n\u001b[32m    258\u001b[39m         \u001b[38;5;28mself\u001b[39m,\n\u001b[32m    259\u001b[39m         \u001b[38;5;28mlist\u001b[39m(args),\n\u001b[32m    260\u001b[39m         \u001b[38;5;28mself\u001b[39m._bound_context.function_call_options.as_attrs(),\n\u001b[32m    261\u001b[39m     )\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\context.py:1688\u001b[39m, in \u001b[36mContext.call_function\u001b[39m\u001b[34m(self, name, tensor_inputs, num_outputs)\u001b[39m\n\u001b[32m   1686\u001b[39m cancellation_context = cancellation.context()\n\u001b[32m   1687\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m cancellation_context \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[32m-> \u001b[39m\u001b[32m1688\u001b[39m   outputs = \u001b[43mexecute\u001b[49m\u001b[43m.\u001b[49m\u001b[43mexecute\u001b[49m\u001b[43m(\u001b[49m\n\u001b[32m   1689\u001b[39m \u001b[43m      \u001b[49m\u001b[43mname\u001b[49m\u001b[43m.\u001b[49m\u001b[43mdecode\u001b[49m\u001b[43m(\u001b[49m\u001b[33;43m\"\u001b[39;49m\u001b[33;43mutf-8\u001b[39;49m\u001b[33;43m\"\u001b[39;49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m   1690\u001b[39m \u001b[43m      \u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[43m=\u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m   1691\u001b[39m \u001b[43m      \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m=\u001b[49m\u001b[43mtensor_inputs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m   1692\u001b[39m \u001b[43m      \u001b[49m\u001b[43mattrs\u001b[49m\u001b[43m=\u001b[49m\u001b[43mattrs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m   1693\u001b[39m \u001b[43m      \u001b[49m\u001b[43mctx\u001b[49m\u001b[43m=\u001b[49m\u001b[38;5;28;43mself\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[32m   1694\u001b[39m \u001b[43m  \u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m   1695\u001b[39m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[32m   1696\u001b[39m   outputs = execute.execute_with_cancellation(\n\u001b[32m   1697\u001b[39m       name.decode(\u001b[33m\"\u001b[39m\u001b[33mutf-8\u001b[39m\u001b[33m\"\u001b[39m),\n\u001b[32m   1698\u001b[39m       num_outputs=num_outputs,\n\u001b[32m   (...)\u001b[39m\u001b[32m   1702\u001b[39m       cancellation_manager=cancellation_context,\n\u001b[32m   1703\u001b[39m   )\n",
+      "\u001b[36mFile \u001b[39m\u001b[32mD:\\PROJECT TA\\.venv\\Lib\\site-packages\\tensorflow\\python\\eager\\execute.py:53\u001b[39m, in \u001b[36mquick_execute\u001b[39m\u001b[34m(op_name, num_outputs, inputs, attrs, ctx, name)\u001b[39m\n\u001b[32m     51\u001b[39m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[32m     52\u001b[39m   ctx.ensure_initialized()\n\u001b[32m---> \u001b[39m\u001b[32m53\u001b[39m   tensors = \u001b[43mpywrap_tfe\u001b[49m\u001b[43m.\u001b[49m\u001b[43mTFE_Py_Execute\u001b[49m\u001b[43m(\u001b[49m\u001b[43mctx\u001b[49m\u001b[43m.\u001b[49m\u001b[43m_handle\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdevice_name\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mop_name\u001b[49m\u001b[43m,\u001b[49m\n\u001b[32m     54\u001b[39m \u001b[43m                                      \u001b[49m\u001b[43minputs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mattrs\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnum_outputs\u001b[49m\u001b[43m)\u001b[49m\n\u001b[32m     55\u001b[39m \u001b[38;5;28;01mexcept\u001b[39;00m core._NotOkStatusException \u001b[38;5;28;01mas\u001b[39;00m e:\n\u001b[32m     56\u001b[39m   \u001b[38;5;28;01mif\u001b[39;00m name \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n",
+      "\u001b[31mKeyboardInterrupt\u001b[39m: "
      ]
     }
    ],
@@ -1051,48 +1016,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": null,
    "id": "0c34bf89",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "================================================================================\n",
-      "📈 EVALUATING MODEL ON TEST SET\n",
-      "================================================================================\n",
-      "\n",
-      "📊 Computing predictions...\n",
-      "✓ Predictions computed\n",
-      "\n",
-      "⏱️  Computing metrics...\n",
-      "================================================================================\n",
-      "Test Loss:     0.0002\n",
-      "Test Accuracy: 1.0000 (100.00%)\n",
-      "================================================================================\n",
-      "\n",
-      "📋 Generating detailed classification report...\n",
-      "--------------------------------------------------------------------------------\n",
-      "                 precision    recall  f1-score   support\n",
-      "\n",
-      "Bacterialblight       1.00      1.00      1.00       241\n",
-      "      Brownspot       1.00      1.00      1.00       243\n",
-      "       Leafsmut       1.00      1.00      1.00       219\n",
-      "\n",
-      "       accuracy                           1.00       703\n",
-      "      macro avg       1.00      1.00      1.00       703\n",
-      "   weighted avg       1.00      1.00      1.00       703\n",
-      "\n",
-      "--------------------------------------------------------------------------------\n",
-      "\n",
-      "✓ Overall Accuracy: 1.0000\n",
-      "================================================================================\n",
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# Evaluate on test set\n",
     "print(\"\\n\" + \"=\" * 80)\n",
@@ -1136,65 +1063,10 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": null,
    "id": "586213e5",
    "metadata": {},
-   "outputs": [
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "🔢 Generating confusion matrix...\n",
-      "--------------------------------------------------------------------------------\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Processing:   0%|                                                             | 0/2 [00:00<?, ?it/s]"
-     ]
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 1000x800 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "Processing: 100%|█████████████████████████████████████████████████████| 2/2 [00:00<00:00,  4.20it/s]"
-     ]
-    },
-    {
-     "name": "stdout",
-     "output_type": "stream",
-     "text": [
-      "\n",
-      "📊 Confusion Matrix Details:\n",
-      "--------------------------------------------------------------------------------\n",
-      "  Bacterialblight     :  241/ 241 correct (100.0%)\n",
-      "  Brownspot           :  243/ 243 correct (100.0%)\n",
-      "  Leafsmut            :  219/ 219 correct (100.0%)\n",
-      "--------------------------------------------------------------------------------\n",
-      "✓ Confusion matrix completed!\n",
-      "\n"
-     ]
-    },
-    {
-     "name": "stderr",
-     "output_type": "stream",
-     "text": [
-      "\n"
-     ]
-    }
-   ],
+   "outputs": [],
    "source": [
     "# Plot confusion matrix\n",
     "print(\"🔢 Generating confusion matrix...\")\n",
@@ -1660,9 +1532,9 @@
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "base",
+   "display_name": "Python (TA-Project)",
    "language": "python",
-   "name": "python3"
+   "name": "venv_ta"
   },
   "language_info": {
    "codemirror_mode": {
@@ -1674,7 +1546,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.12.5"
+   "version": "3.11.5"
   }
  },
  "nbformat": 4,
diff --git a/setup.bat b/setup.bat
index 46e77ca..7554e3f 100644
--- a/setup.bat
+++ b/setup.bat
@@ -44,15 +44,14 @@ echo ========================================
 echo.
 echo Langkah selanjutnya:
 echo.
-echo 1. Buka Terminal 1 dan jalankan Python API:
-echo    cd "rice leaf diseases dataset"
-echo    python api_server.py
-echo.
-echo 2. Buka Terminal 2 dan jalankan Laravel:
+echo 1. Buka Terminal 1 dan jalankan Laravel:
 echo    cd "web_TA"
 echo    php artisan serve
 echo.
-echo 3. Test API:
+echo 2. Test API:
 echo    python test_api.py
 echo.
+echo 3. Pastikan Python environment memiliki dependency inferensi:
+echo    pip install -r "rice leaf diseases dataset\requirements_api.txt"
+echo.
 pause
diff --git a/setup.sh b/setup.sh
index 220bd66..5165466 100644
--- a/setup.sh
+++ b/setup.sh
@@ -25,9 +25,9 @@ fi
 # Check model files
 echo ""
 echo "Cek file model..."
-if [ -f "rice_leaf_diseases dataset/rice_leaf_disease_model.keras" ]; then
+if [ -f "rice leaf diseases dataset/rice_leaf_disease_model.keras" ]; then
     echo "[OK] Model keras ditemukan"
-elif [ -f "rice_leaf_diseases dataset/rice_leaf_disease_model.h5" ]; then
+elif [ -f "rice leaf diseases dataset/rice_leaf_disease_model.h5" ]; then
     echo "[OK] Model h5 ditemukan"
 else
     echo "[WARNING] File model tidak ditemukan"
@@ -41,14 +41,13 @@ echo "========================================"
 echo ""
 echo "Langkah selanjutnya:"
 echo ""
-echo "1. Buka Terminal 1 dan jalankan Python API:"
-echo "   cd \"rice leaf diseases dataset\""
-echo "   python3 api_server.py"
-echo ""
-echo "2. Buka Terminal 2 dan jalankan Laravel:"
+echo "1. Buka Terminal 1 dan jalankan Laravel:"
 echo "   cd web_TA"
 echo "   php artisan serve"
 echo ""
-echo "3. Test API:"
+echo "2. Test API:"
 echo "   python3 test_api.py"
 echo ""
+echo "3. Pastikan dependency inferensi Python sudah terpasang:"
+echo "   pip3 install -r \"rice leaf diseases dataset/requirements_api.txt\""
+echo ""
diff --git a/test_api.py b/test_api.py
index b6ff00b..fa91bbb 100644
--- a/test_api.py
+++ b/test_api.py
@@ -3,59 +3,60 @@ Simple test script untuk menguji API Classification
 """
 
 import requests
-import base64
 import json
 from pathlib import Path
 
 # Konfigurasi
 LARAVEL_API_BASE = "http://127.0.0.1:8000/api/classification"
-PYTHON_API_BASE = "http://127.0.0.1:5000"
 
-def test_python_api_health():
-    """Test health check Python API"""
+def test_laravel_api_health():
+    """Test health check API klasifikasi di Laravel"""
     print("\n" + "="*60)
-    print("TEST 1: Python API Health Check")
+    print("TEST 1: Laravel Classification Health Check")
     print("="*60)
     
     try:
-        response = requests.get(f"{PYTHON_API_BASE}/health", timeout=5)
+        response = requests.get(f"{LARAVEL_API_BASE}/health", timeout=15)
         
         if response.status_code == 200:
             data = response.json()
-            print("✅ Python API berhasil dihubungi!")
-            print(f"   Status: {data['status']}")
-            print(f"   Model Loaded: {data['model_loaded']}")
-            print(f"   Classes: {', '.join(data['classes'])}")
+            print("✅ Laravel API berhasil dihubungi!")
+            print(f"   Success: {data.get('success')}")
+            print(f"   Message: {data.get('message')}")
+            model_info = data.get('model_info', {})
+            print(f"   Model Loaded: {model_info.get('model_loaded')}")
+            print(f"   Classes: {', '.join(model_info.get('classes', []))}")
             return True
         else:
             print(f"❌ API responded with status {response.status_code}")
+            print(f"   Response: {response.text}")
             return False
             
     except requests.exceptions.ConnectionError:
-        print("❌ Tidak dapat menghubungi Python API")
-        print(f"   Pastikan server berjalan di {PYTHON_API_BASE}")
+        print("❌ Tidak dapat menghubungi Laravel API")
+        print("   Pastikan server berjalan di http://127.0.0.1:8000")
         return False
     except Exception as e:
         print(f"❌ Error: {str(e)}")
         return False
 
 
-def test_laravel_api_connection():
-    """Test connection via Laravel API"""
+def test_laravel_api_info():
+    """Test endpoint info model via Laravel API"""
     print("\n" + "="*60)
-    print("TEST 2: Laravel API Connection Test")
+    print("TEST 2: Laravel Model Info")
     print("="*60)
     
     try:
-        response = requests.get(f"{LARAVEL_API_BASE}/test", timeout=5)
+        response = requests.get(f"{LARAVEL_API_BASE}/info", timeout=15)
         
         if response.status_code == 200:
             data = response.json()
-            print("✅ Laravel API berhasil dihubungi!")
+            print("✅ Endpoint info berhasil diakses!")
             print(f"   Success: {data['success']}")
             print(f"   Message: {data['message']}")
-            if 'model_info' in data:
-                print(f"   Model Info: {json.dumps(data['model_info'], indent=2)}")
+            if 'data' in data:
+                print(f"   Model Info: {json.dumps(data['data'], indent=2)}")
             return True
         else:
             print(f"❌ API responded with status {response.status_code}")
@@ -167,11 +168,9 @@ def main():
     print("# Rice Leaf Disease Classification API - Test Suite")
     print("#" * 60)
     
-    # Test Python API
-    python_ok = test_python_api_health()
-    
-    # Test Laravel API
-    laravel_ok = test_laravel_api_connection()
+    # Test Laravel API health dan info
+    health_ok = test_laravel_api_health()
+    info_ok = test_laravel_api_info()
     
     # Test dengan gambar contoh jika ada
     test_image_paths = [
@@ -202,8 +201,8 @@ def main():
     print("\n" + "="*60)
     print("TEST SUMMARY")
     print("="*60)
-    print(f"Python API:  {'✅ OK' if python_ok else '❌ FAILED'}")
-    print(f"Laravel API: {'✅ OK' if laravel_ok else '❌ FAILED'}")
+    print(f"Laravel Health: {'✅ OK' if health_ok else '❌ FAILED'}")
+    print(f"Laravel Info:   {'✅ OK' if info_ok else '❌ FAILED'}")
     print("\nUntuk hasil lengkap, sediakan file gambar test.\n")
 
 
diff --git a/web_TA/.env.example b/web_TA/.env.example
index c0660ea..6de80cf 100644
--- a/web_TA/.env.example
+++ b/web_TA/.env.example
@@ -62,4 +62,8 @@ AWS_DEFAULT_REGION=us-east-1
 AWS_BUCKET=
 AWS_USE_PATH_STYLE_ENDPOINT=false
 
+PYTHON_EXECUTABLE=python
+PYTHON_CLASSIFIER_SCRIPT=
+RICE_MODEL_DIR=
+
 VITE_APP_NAME="${APP_NAME}"
diff --git a/web_TA/app/Http/Controllers/ClassificationController.php b/web_TA/app/Http/Controllers/ClassificationController.php
index 46addd9..8cda12e 100644
--- a/web_TA/app/Http/Controllers/ClassificationController.php
+++ b/web_TA/app/Http/Controllers/ClassificationController.php
@@ -3,16 +3,20 @@
 namespace App\Http\Controllers;
 
 use Illuminate\Http\Request;
-use Illuminate\Support\Facades\Http;
 use Illuminate\Support\Facades\Storage;
+use Illuminate\Support\Facades\Log;
 use App\Models\Classification;
+use App\Services\PythonClassificationService;
 
 class ClassificationController extends Controller
 {
-    private $pythonApiUrl = 'http://127.0.0.1:5000'; // URL Flask API
+    public function __construct(
+        private readonly PythonClassificationService $classificationService
+    ) {
+    }
 
     /**
-     * Menerima gambar dan mengirim ke Python API untuk klasifikasi
+     * Menerima gambar dan mengklasifikasi melalui service internal Laravel
      * POST /api/classify
      */
     public function classify(Request $request)
@@ -30,36 +34,44 @@ public function classify(Request $request)
             $imageContent = file_get_contents($file->getRealPath());
             $base64Image = base64_encode($imageContent);
 
-            // Kirim request ke Flask API
-            $response = Http::timeout(30)->post($this->pythonApiUrl . '/classify', [
+            // Jalankan klasifikasi melalui service lokal (tanpa Flask API)
+            $result = $this->classificationService->classifyFromBase64([
                 'image' => $base64Image,
                 'filename' => $file->getClientOriginalName(),
             ]);
 
-            // Cek apakah request berhasil
-            if ($response->failed()) {
+            if (!($result['success'] ?? false)) {
                 return response()->json([
                     'success' => false,
                     'message' => 'Gagal menghubungi model classification',
-                    'error' => $response->body()
+                    'error' => $result['message'] ?? 'Unknown error',
                 ], 500);
             }
 
-            $result = $response->json();
-
             // Tambahkan informasi detail tentang penyakit
             $diseaseInfo = $this->getDiseaseInfo($result['predicted_class']);
 
-            // Save to database
-            Classification::create([
-                'filename' => $file->getClientOriginalName(),
-                'predicted_class' => $result['predicted_class'],
-                'confidence' => $result['confidence'],
-                'all_predictions' => $result['all_predictions'],
-                'disease_name' => $diseaseInfo['name'],
-                'severity' => $diseaseInfo['severity'],
-                'notes' => 'Classification without storage',
-            ]);
+            $savedToDatabase = true;
+            $persistenceWarning = null;
+
+            try {
+                Classification::create([
+                    'filename' => $file->getClientOriginalName(),
+                    'predicted_class' => $result['predicted_class'],
+                    'confidence' => $result['confidence'],
+                    'all_predictions' => $result['all_predictions'],
+                    'disease_name' => $diseaseInfo['name'],
+                    'severity' => $diseaseInfo['severity'],
+                    'notes' => 'Classification without storage',
+                ]);
+            } catch (\Throwable $dbException) {
+                $savedToDatabase = false;
+                $persistenceWarning = 'Klasifikasi berhasil, tetapi gagal simpan ke database.';
+                Log::warning('Classification result not persisted', [
+                    'filename' => $file->getClientOriginalName(),
+                    'error' => $dbException->getMessage(),
+                ]);
+            }
 
             return response()->json([
                 'success' => true,
@@ -70,6 +82,8 @@ public function classify(Request $request)
                     'confidence_value' => $result['confidence'],
                     'all_predictions' => $result['all_predictions'],
                     'disease_info' => $diseaseInfo,
+                    'saved_to_database' => $savedToDatabase,
+                    'persistence_warning' => $persistenceWarning,
                     'timestamp' => now(),
                 ]
             ], 200);
@@ -89,7 +103,7 @@ public function classify(Request $request)
     }
 
     /**
-     * Upload gambar dan simpan ke database, kemudian klasifikasi
+        * Upload gambar dan simpan ke database, kemudian klasifikasi
      * POST /api/classify-and-save
      */
     public function classifyAndSave(Request $request)
@@ -109,36 +123,47 @@ public function classifyAndSave(Request $request)
             $imageContent = file_get_contents($file->getRealPath());
             $base64Image = base64_encode($imageContent);
 
-            // Kirim request ke Flask API
-            $response = Http::timeout(30)->post($this->pythonApiUrl . '/classify', [
+            // Jalankan klasifikasi melalui service lokal (tanpa Flask API)
+            $result = $this->classificationService->classifyFromBase64([
                 'image' => $base64Image,
                 'filename' => $file->getClientOriginalName(),
             ]);
 
-            if ($response->failed()) {
+            if (!($result['success'] ?? false)) {
                 // Hapus file yang sudah disimpan jika klasifikasi gagal
                 Storage::disk('public')->delete($storagePath);
                 
                 return response()->json([
                     'success' => false,
                     'message' => 'Gagal menghubungi model classification',
+                    'error' => $result['message'] ?? 'Unknown error',
                 ], 500);
             }
-
-            $result = $response->json();
             $diseaseInfo = $this->getDiseaseInfo($result['predicted_class']);
 
-            // Save to database
-            Classification::create([
-                'image_path' => $storagePath,
-                'filename' => $file->getClientOriginalName(),
-                'predicted_class' => $result['predicted_class'],
-                'confidence' => $result['confidence'],
-                'all_predictions' => $result['all_predictions'],
-                'disease_name' => $diseaseInfo['name'],
-                'severity' => $diseaseInfo['severity'],
-                'notes' => $request->input('notes'),
-            ]);
+            $savedToDatabase = true;
+            $persistenceWarning = null;
+
+            try {
+                Classification::create([
+                    'image_path' => $storagePath,
+                    'filename' => $file->getClientOriginalName(),
+                    'predicted_class' => $result['predicted_class'],
+                    'confidence' => $result['confidence'],
+                    'all_predictions' => $result['all_predictions'],
+                    'disease_name' => $diseaseInfo['name'],
+                    'severity' => $diseaseInfo['severity'],
+                    'notes' => $request->input('notes'),
+                ]);
+            } catch (\Throwable $dbException) {
+                $savedToDatabase = false;
+                $persistenceWarning = 'Gambar berhasil diklasifikasi dan disimpan file, tetapi gagal simpan riwayat ke database.';
+                Log::warning('Classification file stored but DB persist failed', [
+                    'filename' => $file->getClientOriginalName(),
+                    'path' => $storagePath,
+                    'error' => $dbException->getMessage(),
+                ]);
+            }
 
             return response()->json([
                 'success' => true,
@@ -150,6 +175,8 @@ public function classifyAndSave(Request $request)
                     'confidence_value' => $result['confidence'],
                     'all_predictions' => $result['all_predictions'],
                     'disease_info' => $diseaseInfo,
+                    'saved_to_database' => $savedToDatabase,
+                    'persistence_warning' => $persistenceWarning,
                     'notes' => $request->input('notes'),
                     'timestamp' => now(),
                 ]
@@ -170,34 +197,143 @@ public function classifyAndSave(Request $request)
     }
 
     /**
-     * Test koneksi ke Python API
+     * Klasifikasi gambar dari URL
+     * POST /api/classification/classify-from-url
+     */
+    public function classifyFromUrl(Request $request)
+    {
+        try {
+            $request->validate([
+                'image_url' => 'required|url|max:2048',
+                'notes' => 'nullable|string|max:500',
+                'save' => 'nullable|boolean',
+            ]);
+
+            $imageUrl = $request->input('image_url');
+            $result = $this->classificationService->classifyFromUrl([
+                'image_url' => $imageUrl,
+            ]);
+
+            if (!($result['success'] ?? false)) {
+                return response()->json([
+                    'success' => false,
+                    'message' => $result['message'] ?? 'Gagal memproses gambar dari URL',
+                ], 400);
+            }
+
+            $diseaseInfo = $this->getDiseaseInfo($result['predicted_class']);
+
+            if ($request->boolean('save')) {
+                $urlPath = parse_url($imageUrl, PHP_URL_PATH);
+                $filename = is_string($urlPath) && $urlPath !== ''
+                    ? basename($urlPath)
+                    : 'from_url_image';
+
+                Classification::create([
+                    'filename' => $filename,
+                    'predicted_class' => $result['predicted_class'],
+                    'confidence' => $result['confidence'],
+                    'all_predictions' => $result['all_predictions'],
+                    'disease_name' => $diseaseInfo['name'],
+                    'severity' => $diseaseInfo['severity'],
+                    'notes' => $request->input('notes', 'Classification from URL'),
+                ]);
+            }
+
+            return response()->json([
+                'success' => true,
+                'message' => 'Klasifikasi dari URL berhasil',
+                'data' => [
+                    'url' => $imageUrl,
+                    'predicted_class' => $result['predicted_class'],
+                    'confidence' => round($result['confidence'] * 100, 2) . '%',
+                    'confidence_value' => $result['confidence'],
+                    'all_predictions' => $result['all_predictions'],
+                    'disease_info' => $diseaseInfo,
+                    'timestamp' => now(),
+                ]
+            ], 200);
+        } catch (\Illuminate\Validation\ValidationException $e) {
+            return response()->json([
+                'success' => false,
+                'message' => 'Validasi gagal',
+                'errors' => $e->errors()
+            ], 422);
+        } catch (\Exception $e) {
+            return response()->json([
+                'success' => false,
+                'message' => 'Terjadi kesalahan: ' . $e->getMessage(),
+            ], 500);
+        }
+    }
+
+    /**
+     * Health check klasifikasi service
+     * GET /api/classification/health
+     */
+    public function health()
+    {
+        try {
+            $result = $this->classificationService->health();
+
+            if ($result['status'] !== 'ok') {
+                return response()->json([
+                    'success' => false,
+                    'message' => $result['message'] ?? 'Model tidak siap',
+                    'model_info' => $result,
+                ], 503);
+            }
+
+            return response()->json([
+                'success' => true,
+                'message' => 'Koneksi ke model classification berhasil',
+                'model_info' => $result,
+            ], 200);
+        } catch (\Exception $e) {
+            return response()->json([
+                'success' => false,
+                'message' => 'Tidak dapat melakukan health check model: ' . $e->getMessage(),
+            ], 500);
+        }
+    }
+
+    /**
+     * Informasi model klasifikasi
+     * GET /api/classification/info
+     */
+    public function info()
+    {
+        try {
+            $result = $this->classificationService->info();
+
+            if (!($result['model_loaded'] ?? false)) {
+                return response()->json([
+                    'success' => false,
+                    'message' => $result['message'] ?? 'Model tidak tersedia',
+                    'data' => $result,
+                ], 503);
+            }
+
+            return response()->json([
+                'success' => true,
+                'message' => 'Informasi model berhasil diambil',
+                'data' => $result,
+            ], 200);
+        } catch (\Exception $e) {
+            return response()->json([
+                'success' => false,
+                'message' => 'Gagal mengambil informasi model: ' . $e->getMessage(),
+            ], 500);
+        }
+    }
+
+    /**
+     * Backward-compatible endpoint test (alias untuk health)
      * GET /api/classification/test
      */
     public function testConnection()
     {
-        try {
-            $response = Http::timeout(10)->post($this->pythonApiUrl . '/health', []);
-
-            if ($response->successful()) {
-                return response()->json([
-                    'success' => true,
-                    'message' => 'Koneksi ke model API berhasil',
-                    'model_info' => $response->json()
-                ], 200);
-            }
-
-            return response()->json([
-                'success' => false,
-                'message' => 'Model API tidak merespons dengan benar'
-            ], 500);
-
-        } catch (\Exception $e) {
-            return response()->json([
-                'success' => false,
-                'message' => 'Tidak dapat menghubungi model API: ' . $e->getMessage(),
-                'hint' => 'Pastikan server Python API sudah berjalan di ' . $this->pythonApiUrl
-            ], 500);
-        }
+        return $this->health();
     }
 
     /**
diff --git a/web_TA/app/Services/PythonClassificationService.php b/web_TA/app/Services/PythonClassificationService.php
new file mode 100644
index 0000000..906cbb9
--- /dev/null
+++ b/web_TA/app/Services/PythonClassificationService.php
@@ -0,0 +1,130 @@
+<?php
+
+namespace App\Services;
+
+use RuntimeException;
+use Symfony\Component\Process\Process;
+
+class PythonClassificationService
+{
+    private string $pythonExecutable;
+    private string $scriptPath;
+    private string $modelDirectory;
+
+    public function __construct()
+    {
+        $this->pythonExecutable = env('PYTHON_EXECUTABLE', 'python');
+        $this->scriptPath = env('PYTHON_CLASSIFIER_SCRIPT', base_path('scripts/rice_inference.py'));
+        $this->modelDirectory = env('RICE_MODEL_DIR', base_path('../rice leaf diseases dataset'));
+    }
+
+    public function classifyFromBase64(array $payload): array
+    {
+        return $this->runAction('classify', $payload, 120);
+    }
+
+    public function classifyFromUrl(array $payload): array
+    {
+        return $this->runAction('classify-from-url', $payload, 120);
+    }
+
+    public function health(): array
+    {
+        return $this->runAction('health');
+    }
+
+    public function info(): array
+    {
+        return $this->runAction('info');
+    }
+
+    private function runAction(string $action, array $payload = [], int $timeout = 60): array
+    {
+        if (!is_file($this->scriptPath)) {
+            throw new RuntimeException("Script classifier tidak ditemukan di {$this->scriptPath}");
+        }
+
+        $command = [
+            $this->pythonExecutable,
+            $this->scriptPath,
+            $action,
+            '--model-dir',
+            $this->modelDirectory,
+        ];
+
+        $process = new Process($command, base_path(), $this->buildProcessEnvironment());
+        $process->setTimeout($timeout);
+        $process->setInput(json_encode($payload, JSON_UNESCAPED_SLASHES));
+        $process->run();
+
+        if (!$process->isSuccessful()) {
+            $stderr = trim($process->getErrorOutput());
+            $stdout = trim($process->getOutput());
+            throw new RuntimeException($stderr !== '' ? $stderr : ($stdout !== '' ? $stdout : 'Gagal menjalankan proses inferensi Python'));
+        }
+
+        $output = trim($process->getOutput());
+        if ($output === '') {
+            throw new RuntimeException('Proses inferensi Python tidak mengembalikan output.');
+        }
+
+        $decoded = json_decode($output, true);
+        if (!is_array($decoded)) {
+            throw new RuntimeException('Output inferensi Python bukan JSON yang valid.');
+        }
+
+        return $decoded;
+    }
+
+    private function buildProcessEnvironment(): array
+    {
+        $environment = array_merge($_SERVER, $_ENV);
+
+        // Prevent Python from using conflicting host-level overrides.
+        unset($environment['PYTHONHOME'], $environment['PYTHONPATH']);
+
+        $pythonDir = dirname($this->pythonExecutable);
+        $currentPath = getenv('PATH') ?: ($environment['PATH'] ?? '');
+        $environment['PATH'] = $pythonDir . PATH_SEPARATOR . $currentPath;
+
+        if (!isset($environment['SystemRoot']) || $environment['SystemRoot'] === '') {
+            $environment['SystemRoot'] = getenv('SystemRoot') ?: 'C:\\Windows';
+        }
+
+        if (!isset($environment['WINDIR']) || $environment['WINDIR'] === '') {
+            $environment['WINDIR'] = getenv('WINDIR') ?: 'C:\\Windows';
+        }
+
+        $fallbackUserProfile = getenv('USERPROFILE') ?: ('C:\\Users\\' . (getenv('USERNAME') ?: 'Public'));
+
+        if (!isset($environment['USERPROFILE']) || $environment['USERPROFILE'] === '') {
+            $environment['USERPROFILE'] = $fallbackUserProfile;
+        }
+
+        if (!isset($environment['HOMEDRIVE']) || $environment['HOMEDRIVE'] === '') {
+            $environment['HOMEDRIVE'] = getenv('HOMEDRIVE') ?: substr($fallbackUserProfile, 0, 2);
+        }
+
+        if (!isset($environment['HOMEPATH']) || $environment['HOMEPATH'] === '') {
+            $environment['HOMEPATH'] = getenv('HOMEPATH') ?: substr($fallbackUserProfile, 2);
+        }
+
+        if (!isset($environment['APPDATA']) || $environment['APPDATA'] === '') {
+            $environment['APPDATA'] = getenv('APPDATA') ?: ($fallbackUserProfile . '\\AppData\\Roaming');
+        }
+
+        if (!isset($environment['LOCALAPPDATA']) || $environment['LOCALAPPDATA'] === '') {
+            $environment['LOCALAPPDATA'] = getenv('LOCALAPPDATA') ?: ($fallbackUserProfile . '\\AppData\\Local');
+        }
+
+        if (!isset($environment['TEMP']) || $environment['TEMP'] === '') {
+            $environment['TEMP'] = getenv('TEMP') ?: ($fallbackUserProfile . '\\AppData\\Local\\Temp');
+        }
+
+        if (!isset($environment['TMP']) || $environment['TMP'] === '') {
+            $environment['TMP'] = getenv('TMP') ?: $environment['TEMP'];
+        }
+
+        return $environment;
+    }
+}
diff --git a/web_TA/routes/api.php b/web_TA/routes/api.php
index 0664ce1..59d2bd3 100644
--- a/web_TA/routes/api.php
+++ b/web_TA/routes/api.php
@@ -15,11 +15,16 @@
 // Classification endpoints
 Route::prefix('classification')->group(function () {
     
-    // Test koneksi ke Python API
+    // Health check klasifikasi (alias test untuk backward compatibility)
+    Route::get('/health', [ClassificationController::class, 'health']);
+    Route::get('/info', [ClassificationController::class, 'info']);
     Route::get('/test', [ClassificationController::class, 'testConnection']);
     
     // Klasifikasi gambar (hanya analisis)
     Route::post('/classify', [ClassificationController::class, 'classify']);
+
+    // Klasifikasi gambar dari URL
+    Route::post('/classify-from-url', [ClassificationController::class, 'classifyFromUrl']);
     
     // Klasifikasi dan simpan gambar
     Route::post('/classify-and-save', [ClassificationController::class, 'classifyAndSave']);
diff --git a/web_TA/scripts/rice_inference.py b/web_TA/scripts/rice_inference.py
new file mode 100644
index 0000000..e950007
--- /dev/null
+++ b/web_TA/scripts/rice_inference.py
@@ -0,0 +1,293 @@
+"""
+Python CLI untuk inferensi model rice leaf disease.
+Dipanggil langsung dari Laravel, tanpa Flask API.
+"""
+
+import argparse
+import base64
+import json
+import os
+import sys
+from io import BytesIO
+
+import numpy as np
+import requests
+from PIL import Image
+
+
+CLASS_NAMES = ["Bacterialblight", "Brownspot", "Leafsmut"]
+IMG_SIZE = (224, 224)
+
+try:
+    from tensorflow import keras
+    TENSORFLOW_AVAILABLE = True
+    TENSORFLOW_IMPORT_ERROR = None
+except Exception:
+    keras = None
+    TENSORFLOW_AVAILABLE = False
+    TENSORFLOW_IMPORT_ERROR = str(sys.exc_info()[1])
+
+MODEL = None
+MODEL_PATH = None
+
+
+def _read_json_input() -> dict:
+    raw = sys.stdin.read().strip()
+    if not raw:
+        return {}
+    try:
+        data = json.loads(raw)
+        return data if isinstance(data, dict) else {}
+    except json.JSONDecodeError:
+        return {}
+
+
+def _emit(data: dict, exit_code: int = 0) -> None:
+    print(json.dumps(data, ensure_ascii=True))
+    raise SystemExit(exit_code)
+
+
+def _find_model_file(model_dir: str) -> str | None:
+    candidates = [
+        "rice_leaf_disease_model.keras",
+        "rice_leaf_disease_model.h5",
+        "rice_leaf_disease_model.json",
+    ]
+
+    for file_name in candidates:
+        full_path = os.path.join(model_dir, file_name)
+        if os.path.isfile(full_path):
+            return full_path
+    return None
+
+
+def _load_model(model_dir: str):
+    global MODEL, MODEL_PATH
+
+    if MODEL is not None:
+        return MODEL
+
+    if not TENSORFLOW_AVAILABLE:
+        raise RuntimeError("TensorFlow tidak tersedia pada environment Python ini.")
+
+    model_file = _find_model_file(model_dir)
+    if model_file is None:
+        raise RuntimeError(
+            "Model tidak ditemukan. Pastikan salah satu file ini ada: "
+            "rice_leaf_disease_model.keras, rice_leaf_disease_model.h5, rice_leaf_disease_model.json"
+        )
+
+    MODEL_PATH = model_file
+
+    if model_file.endswith(".keras") or model_file.endswith(".h5"):
+        MODEL = keras.models.load_model(model_file)
+        return MODEL
+
+    with open(model_file, "r", encoding="utf-8") as f:
+        model_json = f.read()
+
+    MODEL = keras.models.model_from_json(model_json)
+
+    weights_base = model_file.replace(".json", "")
+    weight_candidates = [
+        f"{weights_base}.h5",
+        f"{weights_base}_weights.h5",
+    ]
+
+    for weights_file in weight_candidates:
+        if os.path.isfile(weights_file):
+            MODEL.load_weights(weights_file)
+            return MODEL
+
+    raise RuntimeError("Model JSON ditemukan, tetapi file weights tidak ditemukan.")
+
+
+def _preprocess_image(image_bytes: bytes):
+    image = Image.open(BytesIO(image_bytes))
+    if image.mode != "RGB":
+        image = image.convert("RGB")
+
+    try:
+        resample_filter = Image.Resampling.LANCZOS
+    except AttributeError:
+        resample_filter = Image.LANCZOS
+
+    image = image.resize(IMG_SIZE, resample_filter)
+
+    img_array = np.array(image, dtype="float32") / 255.0
+    img_array = np.expand_dims(img_array, axis=0)
+    return img_array
+
+
+def _predict(image_bytes: bytes, model_dir: str) -> dict:
+    model = _load_model(model_dir)
+
+    img_array = _preprocess_image(image_bytes)
+    predictions = model.predict(img_array, verbose=0)
+
+    predicted_idx = int(np.argmax(predictions[0]))
+    predicted_class = CLASS_NAMES[predicted_idx]
+    confidence = float(predictions[0][predicted_idx])
+
+    all_predictions = {
+        class_name: float(predictions[0][idx])
+        for idx, class_name in enumerate(CLASS_NAMES)
+    }
+
+    return {
+        "success": True,
+        "predicted_class": predicted_class,
+        "confidence": confidence,
+        "all_predictions": all_predictions,
+        "model_path": MODEL_PATH,
+    }
+
+
+def action_classify(model_dir: str) -> None:
+    payload = _read_json_input()
+    image_base64 = payload.get("image")
+
+    if not image_base64 or not isinstance(image_base64, str):
+        _emit({"success": False, "message": "Field 'image' (base64) diperlukan"}, 0)
+
+    try:
+        image_bytes = base64.b64decode(image_base64)
+        result = _predict(image_bytes, model_dir)
+        _emit(result, 0)
+    except Exception as e:
+        _emit({"success": False, "message": str(e)}, 0)
+
+
+def action_classify_from_url(model_dir: str) -> None:
+    payload = _read_json_input()
+    image_url = payload.get("image_url")
+
+    if not image_url or not isinstance(image_url, str):
+        _emit({"success": False, "message": "Field 'image_url' diperlukan"}, 0)
+
+    try:
+        response = requests.get(image_url, timeout=10)
+        if response.status_code != 200:
+            _emit({"success": False, "message": "Gagal download image dari URL"}, 0)
+
+        result = _predict(response.content, model_dir)
+        result["url"] = image_url
+        _emit(result, 0)
+    except Exception as e:
+        _emit({"success": False, "message": str(e)}, 0)
+
+
+def action_health(model_dir: str) -> None:
+    if not TENSORFLOW_AVAILABLE:
+        message = "TensorFlow tidak tersedia"
+        if TENSORFLOW_IMPORT_ERROR:
+            message = f"TensorFlow tidak tersedia: {TENSORFLOW_IMPORT_ERROR}"
+
+        _emit(
+            {
+                "status": "error",
+                "message": message,
+                "model_loaded": False,
+                "python_executable": sys.executable,
+                "classes": CLASS_NAMES,
+            },
+            0,
+        )
+
+    try:
+        model = _load_model(model_dir)
+        _emit(
+            {
+                "status": "ok",
+                "message": "Model siap digunakan",
+                "model_loaded": True,
+                "model_path": MODEL_PATH,
+                "python_executable": sys.executable,
+                "classes": CLASS_NAMES,
+                "input_shape": str(model.input_shape),
+            },
+            0,
+        )
+    except Exception as e:
+        _emit(
+            {
+                "status": "error",
+                "message": str(e),
+                "model_loaded": False,
+                "python_executable": sys.executable,
+                "classes": CLASS_NAMES,
+            },
+            0,
+        )
+
+
+def action_info(model_dir: str) -> None:
+    if not TENSORFLOW_AVAILABLE:
+        message = "TensorFlow tidak tersedia"
+        if TENSORFLOW_IMPORT_ERROR:
+            message = f"TensorFlow tidak tersedia: {TENSORFLOW_IMPORT_ERROR}"
+
+        _emit(
+            {
+                "model_loaded": False,
+                "message": message,
+                "python_executable": sys.executable,
+                "classes": CLASS_NAMES,
+                "number_of_classes": len(CLASS_NAMES),
+            },
+            0,
+        )
+
+    try:
+        model = _load_model(model_dir)
+        _emit(
+            {
+                "model_loaded": True,
+                "model_path": MODEL_PATH,
+                "python_executable": sys.executable,
+                "classes": CLASS_NAMES,
+                "number_of_classes": len(CLASS_NAMES),
+                "input_shape": str(model.input_shape),
+                "number_of_layers": len(model.layers),
+                "total_parameters": int(model.count_params()),
+            },
+            0,
+        )
+    except Exception as e:
+        _emit(
+            {
+                "model_loaded": False,
+                "message": str(e),
+                "python_executable": sys.executable,
+                "classes": CLASS_NAMES,
+                "number_of_classes": len(CLASS_NAMES),
+            },
+            0,
+        )
+
+
+def main() -> None:
+    parser = argparse.ArgumentParser(description="Rice leaf disease inference CLI")
+    parser.add_argument(
+        "action",
+        choices=["classify", "classify-from-url", "health", "info"],
+    )
+    parser.add_argument(
+        "--model-dir",
+        required=True,
+        help="Direktori tempat file model berada",
+    )
+    args = parser.parse_args()
+
+    if args.action == "classify":
+        action_classify(args.model_dir)
+    if args.action == "classify-from-url":
+        action_classify_from_url(args.model_dir)
+    if args.action == "health":
+        action_health(args.model_dir)
+    if args.action == "info":
+        action_info(args.model_dir)
+
+
+if __name__ == "__main__":
+    main()
diff --git a/web_TA/~/.keras/keras.json b/web_TA/~/.keras/keras.json
new file mode 100644
index 0000000..bc2cae3
--- /dev/null
+++ b/web_TA/~/.keras/keras.json
@@ -0,0 +1,6 @@
+{
+    "floatx": "float32",
+    "epsilon": 1e-07,
+    "backend": "tensorflow",
+    "image_data_format": "channels_last"
+}
\ No newline at end of file