{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "3a61e373-68f2-4556-882a-1bf9a773cf80",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TensorFlow version: 2.20.0\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 1. IMPORT LIBRARY YANG DIBUTUHKAN\n",
    "# =============================================================================\n",
    "import tensorflow as tf\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras import layers\n",
    "from tensorflow.keras.models import Sequential\n",
    "import matplotlib.pyplot as plt\n",
    "import json\n",
    "import os\n",
    "\n",
    "print(\"TensorFlow version:\", tf.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ef903114-1810-4dd9-a1b9-db7c5b22f41d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# =============================================================================\n",
    "# 2. TENTUKAN PARAMETER UTAMA\n",
    "# =============================================================================\n",
    "# Ukuran gambar yang akan dimasukkan ke model\n",
    "IMAGE_SIZE = (128, 128) \n",
    "# Jumlah gambar yang diproses dalam satu waktu\n",
    "BATCH_SIZE = 32 \n",
    "# Path ke folder dataset utama Anda\n",
    "DATASET_DIR = '../Buah Dan Sayur' \n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "1abf6b66-19cb-4fdd-ba0f-ca4423b0bcce",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Memuat gambar dari direktori: ../Buah Dan Sayur\n",
      "Found 15001 files belonging to 15 classes.\n",
      "Using 10501 files for training.\n",
      "Found 15001 files belonging to 15 classes.\n",
      "Using 4500 files for validation.\n",
      "\n",
      "Jumlah Batch Training  : 329\n",
      "Jumlah Batch Validation: 70\n",
      "Jumlah Batch Test      : 71\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 3. MEMUAT DAN MEMBAGI DATASET (VERSI 3-WAY SPLIT)\n",
    "# =============================================================================\n",
    "print(f\"Memuat gambar dari direktori: {DATASET_DIR}\")\n",
    "\n",
    "# --- Langkah 1: Memuat Data Training (70%) ---\n",
    "# Kita mengambil 70% untuk training, sisanya 30% akan digunakan untuk Validasi + Test\n",
    "train_dataset = tf.keras.utils.image_dataset_from_directory(\n",
    "    DATASET_DIR,\n",
    "    validation_split=0.3,       # 30% disisihkan untuk Validation & Test\n",
    "    subset=\"training\",\n",
    "    seed=123,\n",
    "    image_size=IMAGE_SIZE,\n",
    "    batch_size=BATCH_SIZE\n",
    ")\n",
    "\n",
    "# --- Langkah 2: Memuat Data Sisa (30%) ---\n",
    "# Dataset sementara ini berisi campuran data Validasi dan Test\n",
    "val_and_test_dataset = tf.keras.utils.image_dataset_from_directory(\n",
    "    DATASET_DIR,\n",
    "    validation_split=0.3,\n",
    "    subset=\"validation\",        # Ini mengambil 30% sisa data\n",
    "    seed=123,\n",
    "    image_size=IMAGE_SIZE,\n",
    "    batch_size=BATCH_SIZE\n",
    ")\n",
    "\n",
    "# --- Langkah 3: Memisahkan Validasi (15%) dan Test (15%) ---\n",
    "# Menghitung jumlah batch dalam dataset sisa\n",
    "val_batches = tf.data.experimental.cardinality(val_and_test_dataset)\n",
    "\n",
    "# Mengambil setengah bagian pertama untuk Validation (15%)\n",
    "validation_dataset = val_and_test_dataset.take(val_batches // 2)\n",
    "\n",
    "# Mengambil setengah bagian sisanya untuk Test (15%)\n",
    "test_dataset = val_and_test_dataset.skip(val_batches // 2)\n",
    "\n",
    "print(f\"\\nJumlah Batch Training  : {tf.data.experimental.cardinality(train_dataset)}\")\n",
    "print(f\"Jumlah Batch Validation: {tf.data.experimental.cardinality(validation_dataset)}\")\n",
    "print(f\"Jumlah Batch Test      : {tf.data.experimental.cardinality(test_dataset)}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "c0b9ba33-5baf-4385-873e-b2feb35005e8",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Kelas yang ditemukan: ['Bean', 'Bitter_Gourd', 'Bottle_Gourd', 'Brinjal', 'Broccoli', 'Cabbage', 'Capsicum', 'Carrot', 'Cauliflower', 'Cucumber', 'Papaya', 'Potato', 'Pumpkin', 'Radish', 'Tomato']\n",
      "Nama kelas berhasil disimpan di 'models/class_names.json'\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 4. SIMPAN NAMA KELAS & OPTIMASI DATASET\n",
    "# =============================================================================\n",
    "# Mendapatkan nama kelas dari nama folder\n",
    "class_names = train_dataset.class_names\n",
    "print(\"\\nKelas yang ditemukan:\", class_names)\n",
    "num_classes = len(class_names)\n",
    "\n",
    "# Pastikan folder 'models' ada\n",
    "if not os.path.exists('../models'):\n",
    "    os.makedirs('../models')\n",
    "\n",
    "# Simpan nama kelas ke file JSON untuk digunakan di aplikasi web nanti\n",
    "with open('../models/class_names.json', 'w') as f:\n",
    "    json.dump(class_names, f)\n",
    "print(\"Nama kelas berhasil disimpan di 'models/class_names.json'\")\n",
    "\n",
    "# Optimasi performa pipeline data\n",
    "AUTOTUNE = tf.data.AUTOTUNE\n",
    "train_dataset = train_dataset.cache().shuffle(1000).prefetch(buffer_size=AUTOTUNE)\n",
    "validation_dataset = validation_dataset.cache().prefetch(buffer_size=AUTOTUNE)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "7376068f-3b78-4c70-a72b-3228d2b7d4a1",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Python313\\Lib\\site-packages\\keras\\src\\layers\\preprocessing\\tf_data_layer.py:19: UserWarning: Do not pass an `input_shape`/`input_dim` argument to a layer. When using Sequential models, prefer using an `Input(shape)` object as the first layer in the model instead.\n",
      "  super().__init__(**kwargs)\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 5. MEMBUAT ARSITEKTUR MODEL CNN (VERSI OPTIMASI)\n",
    "# =============================================================================\n",
    "\n",
    "# Lapisan Augmentasi Data untuk membuat model lebih tangguh\n",
    "data_augmentation = keras.Sequential(\n",
    "    [\n",
    "        layers.RandomFlip(\"horizontal\", input_shape=(IMAGE_SIZE[0], IMAGE_SIZE[1], 3)),\n",
    "        layers.RandomRotation(0.1),\n",
    "        layers.RandomZoom(0.1),\n",
    "    ]\n",
    ")\n",
    "\n",
    "# Membangun arsitektur model CNN secara sekuensial\n",
    "model = Sequential([\n",
    "    data_augmentation,\n",
    "    layers.Rescaling(1./255), # Normalisasi piksel ke rentang [0, 1]\n",
    "\n",
    "    # Blok Konvolusi 1\n",
    "    layers.Conv2D(32, 3, padding='same', activation='relu'),\n",
    "    layers.MaxPooling2D(),\n",
    "\n",
    "    # Blok Konvolusi 2\n",
    "    layers.Conv2D(64, 3, padding='same', activation='relu'),\n",
    "    layers.MaxPooling2D(),\n",
    "\n",
    "    # Blok Konvolusi 3\n",
    "    layers.Conv2D(128, 3, padding='same', activation='relu'),\n",
    "    layers.MaxPooling2D(),\n",
    "\n",
    "    layers.Dropout(0.2), # Dropout setelah Lapisan konvolusional\n",
    "    layers.Flatten(),\n",
    "\n",
    "    # === MODIFIKASI DIMULAI DI SINI ===\n",
    "    layers.Dense(128, activation='relu'), # DISARANKAN: Ukuran dikurangi dari 256 menjadi 128\n",
    "    layers.Dropout(0.2),                  # PENAMBAHAN: Dropout tambahan untuk regulasi\n",
    "    # === MODIFIKASI SELESAI DI SINI ===\n",
    "\n",
    "    layers.Dense(num_classes, name=\"outputs\", activation='softmax') # Lapisan output\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "94723c58-cf62-4294-836c-63e137689213",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Arsitektur Model:\n"
     ]
    },
    {
     "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_1\"</span>\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1mModel: \"sequential_1\"\u001b[0m\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "<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",
       "│ sequential (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Sequential</span>)              │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ rescaling (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Rescaling</span>)                │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)        │             <span style=\"color: #00af00; text-decoration-color: #00af00\">896</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\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>)          │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</span>)          │          <span style=\"color: #00af00; text-decoration-color: #00af00\">18,496</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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">64</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\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │          <span style=\"color: #00af00; text-decoration-color: #00af00\">73,856</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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</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\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">16</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">128</span>)         │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ flatten (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Flatten</span>)                    │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">32768</span>)               │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</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\">128</span>)                 │       <span style=\"color: #00af00; text-decoration-color: #00af00\">4,194,432</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\">128</span>)                 │               <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ outputs (<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\">15</span>)                  │           <span style=\"color: #00af00; text-decoration-color: #00af00\">1,935</span> │\n",
       "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n",
       "</pre>\n"
      ],
      "text/plain": [
       "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━┓\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",
       "│ sequential (\u001b[38;5;33mSequential\u001b[0m)              │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m3\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ rescaling (\u001b[38;5;33mRescaling\u001b[0m)                │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m3\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ conv2d (\u001b[38;5;33mConv2D\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m128\u001b[0m, \u001b[38;5;34m32\u001b[0m)        │             \u001b[38;5;34m896\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ max_pooling2d (\u001b[38;5;33mMaxPooling2D\u001b[0m)         │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m32\u001b[0m)          │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ conv2d_1 (\u001b[38;5;33mConv2D\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m, \u001b[38;5;34m64\u001b[0m)          │          \u001b[38;5;34m18,496\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ max_pooling2d_1 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m64\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;34m32\u001b[0m, \u001b[38;5;34m32\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │          \u001b[38;5;34m73,856\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ max_pooling2d_2 (\u001b[38;5;33mMaxPooling2D\u001b[0m)       │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m128\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;34m16\u001b[0m, \u001b[38;5;34m16\u001b[0m, \u001b[38;5;34m128\u001b[0m)         │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ flatten (\u001b[38;5;33mFlatten\u001b[0m)                    │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m32768\u001b[0m)               │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dense (\u001b[38;5;33mDense\u001b[0m)                        │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)                 │       \u001b[38;5;34m4,194,432\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ dropout_1 (\u001b[38;5;33mDropout\u001b[0m)                  │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m128\u001b[0m)                 │               \u001b[38;5;34m0\u001b[0m │\n",
       "├──────────────────────────────────────┼─────────────────────────────┼─────────────────┤\n",
       "│ outputs (\u001b[38;5;33mDense\u001b[0m)                      │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m15\u001b[0m)                  │           \u001b[38;5;34m1,935\u001b[0m │\n",
       "└──────────────────────────────────────┴─────────────────────────────┴─────────────────┘\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">4,289,615</span> (16.36 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m4,289,615\u001b[0m (16.36 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">4,289,615</span> (16.36 MB)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m4,289,615\u001b[0m (16.36 MB)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "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\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> (0.00 B)\n",
       "</pre>\n"
      ],
      "text/plain": [
       "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m0\u001b[0m (0.00 B)\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "from tensorflow.keras.optimizers import Adam # PENAMBAHAN: Import optimizer Adam\n",
    "\n",
    "# =============================================================================\n",
    "# 6. COMPILE MODEL (VERSI OPTIMASI)\n",
    "# =============================================================================\n",
    "\n",
    "# Tentukan Learning Rate yang lebih rendah untuk stabilitas\n",
    "learning_rate_to_use = 0.0001 \n",
    "\n",
    "model.compile(\n",
    "    # MODIFIKASI: Gunakan objek Adam dengan Learning Rate yang dituning\n",
    "    optimizer=Adam(learning_rate=learning_rate_to_use),\n",
    "    \n",
    "    # PERHATIAN: Jika lapisan output Anda menggunakan activation='softmax' (seperti di modifikasi #5), \n",
    "    # maka Anda harus menambahkan 'from_logits=False' atau menghilangkannya (default-nya False).\n",
    "    # Namun, karena Anda menggunakan SparseCategoricalCrossentropy() yang kompatibel dengan output softmax,\n",
    "    # kode ini sudah benar:\n",
    "    loss=tf.keras.losses.SparseCategoricalCrossentropy(),\n",
    "    \n",
    "    metrics=['accuracy']\n",
    ")\n",
    "\n",
    "# Menampilkan ringkasan arsitektur model\n",
    "print(\"\\nArsitektur Model:\")\n",
    "model.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "77d60725-7bb9-42f0-ad60-06d2c5c299b4",
   "metadata": {},
   "outputs": [],
   "source": [
    "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint\n",
    "\n",
    "# Tentukan daftar callbacks\n",
    "early_stop = EarlyStopping(\n",
    "    monitor='val_loss',         # Metrik yang dipantau\n",
    "    patience=5,                 # Jumlah epoch yang ditunggu setelah tidak ada perbaikan\n",
    "    restore_best_weights=True,  # Kembalikan bobot model ke yang terbaik\n",
    "    verbose=1\n",
    ")\n",
    "\n",
    "model_checkpoint = ModelCheckpoint(\n",
    "    filepath='best_cnn_weights.h5', # Nama file untuk menyimpan bobot terbaik\n",
    "    monitor='val_loss',\n",
    "    save_best_only=True,\n",
    "    verbose=1\n",
    ")\n",
    "\n",
    "callbacks_list = [early_stop, model_checkpoint]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "49d84c7d-8ba5-46c0-8ca8-116c72d41525",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Memulai proses training...\n",
      "Epoch 1/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 533ms/step - accuracy: 0.2162 - loss: 2.3720\n",
      "Epoch 1: val_loss improved from None to 1.45690, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m197s\u001b[0m 566ms/step - accuracy: 0.3150 - loss: 2.0781 - val_accuracy: 0.5571 - val_loss: 1.4569\n",
      "Epoch 2/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 526ms/step - accuracy: 0.5205 - loss: 1.4683\n",
      "Epoch 2: val_loss improved from 1.45690 to 1.12152, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m181s\u001b[0m 552ms/step - accuracy: 0.5542 - loss: 1.3635 - val_accuracy: 0.6335 - val_loss: 1.1215\n",
      "Epoch 3/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 528ms/step - accuracy: 0.6187 - loss: 1.1615\n",
      "Epoch 3: val_loss improved from 1.12152 to 0.97365, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m182s\u001b[0m 553ms/step - accuracy: 0.6398 - loss: 1.1014 - val_accuracy: 0.6848 - val_loss: 0.9736\n",
      "Epoch 4/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 523ms/step - accuracy: 0.6881 - loss: 0.9480\n",
      "Epoch 4: val_loss improved from 0.97365 to 0.81808, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m181s\u001b[0m 549ms/step - accuracy: 0.6988 - loss: 0.9249 - val_accuracy: 0.7094 - val_loss: 0.8181\n",
      "Epoch 5/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 510ms/step - accuracy: 0.7341 - loss: 0.8135\n",
      "Epoch 5: val_loss did not improve from 0.81808\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m176s\u001b[0m 534ms/step - accuracy: 0.7395 - loss: 0.7999 - val_accuracy: 0.7089 - val_loss: 0.8544\n",
      "Epoch 6/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 511ms/step - accuracy: 0.7685 - loss: 0.7287\n",
      "Epoch 6: val_loss improved from 0.81808 to 0.66098, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m176s\u001b[0m 535ms/step - accuracy: 0.7683 - loss: 0.7144 - val_accuracy: 0.7688 - val_loss: 0.6610\n",
      "Epoch 7/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 514ms/step - accuracy: 0.7829 - loss: 0.6670\n",
      "Epoch 7: val_loss did not improve from 0.66098\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m177s\u001b[0m 537ms/step - accuracy: 0.7877 - loss: 0.6565 - val_accuracy: 0.7638 - val_loss: 0.6715\n",
      "Epoch 8/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 510ms/step - accuracy: 0.8070 - loss: 0.5994\n",
      "Epoch 8: val_loss improved from 0.66098 to 0.61033, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m176s\u001b[0m 534ms/step - accuracy: 0.8051 - loss: 0.6048 - val_accuracy: 0.7862 - val_loss: 0.6103\n",
      "Epoch 9/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 510ms/step - accuracy: 0.8250 - loss: 0.5446\n",
      "Epoch 9: val_loss improved from 0.61033 to 0.59442, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m176s\u001b[0m 534ms/step - accuracy: 0.8232 - loss: 0.5529 - val_accuracy: 0.8004 - val_loss: 0.5944\n",
      "Epoch 10/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 518ms/step - accuracy: 0.8365 - loss: 0.5186\n",
      "Epoch 10: val_loss did not improve from 0.59442\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m178s\u001b[0m 541ms/step - accuracy: 0.8369 - loss: 0.5230 - val_accuracy: 0.7545 - val_loss: 0.7254\n",
      "Epoch 11/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 55s/step - accuracy: 0.8458 - loss: 0.4789 \n",
      "Epoch 11: val_loss improved from 0.59442 to 0.41276, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m18057s\u001b[0m 55s/step - accuracy: 0.8485 - loss: 0.4754 - val_accuracy: 0.8643 - val_loss: 0.4128\n",
      "Epoch 12/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 723ms/step - accuracy: 0.8528 - loss: 0.4478\n",
      "Epoch 12: val_loss improved from 0.41276 to 0.39406, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m249s\u001b[0m 757ms/step - accuracy: 0.8547 - loss: 0.4437 - val_accuracy: 0.8661 - val_loss: 0.3941\n",
      "Epoch 13/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 714ms/step - accuracy: 0.8683 - loss: 0.4158\n",
      "Epoch 13: val_loss improved from 0.39406 to 0.38393, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m259s\u001b[0m 747ms/step - accuracy: 0.8685 - loss: 0.4156 - val_accuracy: 0.8737 - val_loss: 0.3839\n",
      "Epoch 14/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 671ms/step - accuracy: 0.8761 - loss: 0.3946\n",
      "Epoch 14: val_loss improved from 0.38393 to 0.33691, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m232s\u001b[0m 705ms/step - accuracy: 0.8747 - loss: 0.3988 - val_accuracy: 0.8879 - val_loss: 0.3369\n",
      "Epoch 15/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 665ms/step - accuracy: 0.8765 - loss: 0.3898\n",
      "Epoch 15: val_loss did not improve from 0.33691\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m230s\u001b[0m 699ms/step - accuracy: 0.8799 - loss: 0.3804 - val_accuracy: 0.8670 - val_loss: 0.4146\n",
      "Epoch 16/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 622ms/step - accuracy: 0.8834 - loss: 0.3700\n",
      "Epoch 16: val_loss did not improve from 0.33691\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m213s\u001b[0m 647ms/step - accuracy: 0.8906 - loss: 0.3514 - val_accuracy: 0.8549 - val_loss: 0.4431\n",
      "Epoch 17/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 538ms/step - accuracy: 0.8933 - loss: 0.3410\n",
      "Epoch 17: val_loss improved from 0.33691 to 0.32419, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m185s\u001b[0m 564ms/step - accuracy: 0.8959 - loss: 0.3378 - val_accuracy: 0.8888 - val_loss: 0.3242\n",
      "Epoch 18/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 541ms/step - accuracy: 0.9062 - loss: 0.2966\n",
      "Epoch 18: val_loss did not improve from 0.32419\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m186s\u001b[0m 565ms/step - accuracy: 0.9001 - loss: 0.3178 - val_accuracy: 0.8768 - val_loss: 0.3764\n",
      "Epoch 19/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 544ms/step - accuracy: 0.9063 - loss: 0.3048\n",
      "Epoch 19: val_loss did not improve from 0.32419\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m187s\u001b[0m 568ms/step - accuracy: 0.9043 - loss: 0.3034 - val_accuracy: 0.8607 - val_loss: 0.4274\n",
      "Epoch 20/20\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 591ms/step - accuracy: 0.9067 - loss: 0.2932\n",
      "Epoch 20: val_loss improved from 0.32419 to 0.25759, saving model to best_cnn_weights.keras\n",
      "\u001b[1m329/329\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m203s\u001b[0m 618ms/step - accuracy: 0.9067 - loss: 0.2960 - val_accuracy: 0.9156 - val_loss: 0.2576\n",
      "Restoring model weights from the end of the best epoch: 20.\n",
      "Training selesai.\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 7. MELATIH MODEL (VERSI CEPAT & BERSIH)\n",
    "# =============================================================================\n",
    "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint\n",
    "\n",
    "# 1. Definisi Callbacks\n",
    "early_stop = EarlyStopping(\n",
    "    monitor='val_loss',\n",
    "    patience=5,                 # Stop jika tidak membaik selama 5 epoch berturut-turut\n",
    "    restore_best_weights=True,\n",
    "    verbose=1\n",
    ")\n",
    "\n",
    "# PERBAIKAN 1: Mengubah ekstensi .h5 menjadi .keras agar warning hilang\n",
    "model_checkpoint = ModelCheckpoint(\n",
    "    filepath='best_cnn_weights.keras', \n",
    "    monitor='val_loss',\n",
    "    save_best_only=True,\n",
    "    verbose=1\n",
    ")\n",
    "\n",
    "callbacks_list = [early_stop, model_checkpoint]\n",
    "\n",
    "# 2. Proses Training\n",
    "print(\"\\nMemulai proses training...\")\n",
    "\n",
    "# PERBAIKAN 2: Mengubah kembali ke 20 Epochs agar tidak menunggu terlalu lama\n",
    "epochs = 20 \n",
    "\n",
    "history = model.fit(\n",
    "    train_dataset,\n",
    "    validation_data=validation_dataset,\n",
    "    epochs=epochs,\n",
    "    callbacks=callbacks_list\n",
    ")\n",
    "\n",
    "print(\"Training selesai.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "1d510da6-dba0-4346-a1b3-1383300da737",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 8. VISUALISASI HASIL TRAINING (VERSI AMAN EARLY STOPPING)\n",
    "# =============================================================================\n",
    "acc = history.history['accuracy']\n",
    "val_acc = history.history['val_accuracy']\n",
    "loss = history.history['loss']\n",
    "val_loss = history.history['val_loss']\n",
    "\n",
    "# PERBAIKAN: Gunakan len(acc) agar sesuai dengan jumlah epoch yang benar-benar berjalan\n",
    "epochs_range = range(len(acc))\n",
    "\n",
    "plt.figure(figsize=(12, 5))\n",
    "\n",
    "# Plot Akurasi\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(epochs_range, acc, label='Training Accuracy')\n",
    "plt.plot(epochs_range, val_acc, label='Validation Accuracy')\n",
    "plt.legend(loc='lower right')\n",
    "plt.title('Training and Validation Accuracy')\n",
    "\n",
    "# Plot Loss\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.plot(epochs_range, loss, label='Training Loss')\n",
    "plt.plot(epochs_range, val_loss, label='Validation Loss')\n",
    "plt.legend(loc='upper right')\n",
    "plt.title('Training and Validation Loss')\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "a454e904-7735-430c-bd7e-a9f97edf0f96",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "==================================================\n",
      "    MEMULAI EVALUASI LANJUTAN PADA DATA TEST    \n",
      "==================================================\n",
      "Mengambil data test dan melakukan prediksi...\n",
      "\u001b[1m71/71\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m9s\u001b[0m 120ms/step\n",
      "Kelas berhasil dimuat dari JSON: ['Bean', 'Bitter_Gourd', 'Bottle_Gourd', 'Brinjal', 'Broccoli', 'Cabbage', 'Capsicum', 'Carrot', 'Cauliflower', 'Cucumber', 'Papaya', 'Potato', 'Pumpkin', 'Radish', 'Tomato']\n",
      "\n",
      "[1] UJI GENERALISASI\n",
      "Akurasi pada Data Test (Data Baru): 90.88%\n",
      "\n",
      "[2] LAPORAN KLASIFIKASI\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "        Bean       0.74      0.99      0.85       148\n",
      "Bitter_Gourd       0.85      0.90      0.88       161\n",
      "Bottle_Gourd       0.98      0.92      0.95       167\n",
      "     Brinjal       0.88      0.82      0.85       158\n",
      "    Broccoli       0.86      0.87      0.87       159\n",
      "     Cabbage       0.87      0.82      0.84       142\n",
      "    Capsicum       0.93      0.96      0.94       142\n",
      "      Carrot       0.98      1.00      0.99       165\n",
      " Cauliflower       0.88      0.96      0.92       138\n",
      "    Cucumber       0.96      0.76      0.85       145\n",
      "      Papaya       0.93      0.93      0.93       154\n",
      "      Potato       0.99      0.97      0.98       155\n",
      "     Pumpkin       0.96      0.81      0.88       151\n",
      "      Radish       0.97      0.98      0.98       128\n",
      "      Tomato       0.90      0.93      0.92       147\n",
      "\n",
      "    accuracy                           0.91      2260\n",
      "   macro avg       0.91      0.91      0.91      2260\n",
      "weighted avg       0.91      0.91      0.91      2260\n",
      "\n",
      "\n",
      "[3] VISUALISASI CONFUSION MATRIX\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "[4] VISUALISASI ROC CURVE\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Evaluasi Lanjutan Selesai.\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 9. EVALUASI PERFORMA MODEL SECARA KOMPREHENSIF (VERSI FIX JSON)\n",
    "# =============================================================================\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "from sklearn.metrics import classification_report, confusion_matrix, roc_curve, auc\n",
    "import itertools\n",
    "import tensorflow as tf\n",
    "import json # Penting untuk membaca nama kelas\n",
    "\n",
    "print(\"\\n\" + \"=\"*50)\n",
    "print(\"    MEMULAI EVALUASI LANJUTAN PADA DATA TEST    \")\n",
    "print(\"=\"*50)\n",
    "\n",
    "# --- 1. Persiapan Data (Ekstraksi ke NumPy) ---\n",
    "print(\"Mengambil data test dan melakukan prediksi...\")\n",
    "\n",
    "# Konversi dataset ke numpy array\n",
    "test_images = []\n",
    "test_labels = []\n",
    "\n",
    "for images, labels in test_dataset:\n",
    "    test_images.append(images.numpy())\n",
    "    test_labels.append(labels.numpy())\n",
    "\n",
    "# Menggabungkan batch\n",
    "x_test = np.concatenate(test_images)\n",
    "y_true = np.concatenate(test_labels)\n",
    "\n",
    "# --- 2. Melakukan Prediksi ---\n",
    "y_pred_probs = model.predict(x_test)\n",
    "y_pred = np.argmax(y_pred_probs, axis=1)\n",
    "\n",
    "# === PERBAIKAN DI SINI: Memuat nama kelas dari file JSON ===\n",
    "# Karena train_dataset sudah di-prefetch, kita ambil nama dari file yang disimpan di Bagian #4\n",
    "try:\n",
    "    with open('../models/class_names.json', 'r') as f:\n",
    "        class_names = json.load(f)\n",
    "    print(f\"Kelas berhasil dimuat dari JSON: {class_names}\")\n",
    "except FileNotFoundError:\n",
    "    # Fallback jika file json tidak ketemu (misal path beda), kita coba ambil dari dataset\n",
    "    print(\"Warning: File JSON tidak ditemukan, mencoba mengambil manual...\")\n",
    "    # Urutan folder biasanya alfabetis, pastikan sesuai dengan dataset Anda\n",
    "    import os\n",
    "    class_names = sorted(os.listdir(DATASET_DIR)) \n",
    "    # Hapus file sistem tersembunyi jika ada\n",
    "    class_names = [x for x in class_names if not x.startswith('.')] \n",
    "\n",
    "# --- 3. Uji Generalisasi ---\n",
    "print(\"\\n[1] UJI GENERALISASI\")\n",
    "from sklearn.metrics import accuracy_score\n",
    "test_accuracy = accuracy_score(y_true, y_pred)\n",
    "print(f\"Akurasi pada Data Test (Data Baru): {test_accuracy * 100:.2f}%\")\n",
    "\n",
    "# --- 4. Laporan Klasifikasi ---\n",
    "print(\"\\n[2] LAPORAN KLASIFIKASI\")\n",
    "print(classification_report(y_true, y_pred, target_names=class_names))\n",
    "\n",
    "# --- 5. Visualisasi Confusion Matrix ---\n",
    "print(\"\\n[3] VISUALISASI CONFUSION MATRIX\")\n",
    "\n",
    "def plot_confusion_matrix(cm, classes, normalize=False, title='Confusion Matrix', cmap=plt.cm.Blues):\n",
    "    if normalize:\n",
    "        cm = cm.astype('float') / cm.sum(axis=1)[:, np.newaxis]\n",
    "        title = 'Normalized ' + title\n",
    "    \n",
    "    plt.figure(figsize=(10, 8))\n",
    "    plt.imshow(cm, interpolation='nearest', cmap=cmap)\n",
    "    plt.title(title)\n",
    "    plt.colorbar()\n",
    "    tick_marks = np.arange(len(classes))\n",
    "    plt.xticks(tick_marks, classes, rotation=45, ha=\"right\")\n",
    "    plt.yticks(tick_marks, classes)\n",
    "\n",
    "    fmt = '.2f' if normalize else 'd'\n",
    "    thresh = cm.max() / 2.\n",
    "    for i, j in itertools.product(range(cm.shape[0]), range(cm.shape[1])):\n",
    "        plt.text(j, i, format(cm[i, j], fmt),\n",
    "                 horizontalalignment=\"center\",\n",
    "                 color=\"white\" if cm[i, j] > thresh else \"black\")\n",
    "\n",
    "    plt.ylabel('Label Asli')\n",
    "    plt.xlabel('Label Prediksi')\n",
    "    plt.tight_layout()\n",
    "    plt.show()\n",
    "\n",
    "cm = confusion_matrix(y_true, y_pred)\n",
    "plot_confusion_matrix(cm, classes=class_names, title='Confusion Matrix')\n",
    "\n",
    "# --- 6. Visualisasi ROC Curve ---\n",
    "print(\"\\n[4] VISUALISASI ROC CURVE\")\n",
    "\n",
    "y_true_one_hot = tf.keras.utils.to_categorical(y_true, num_classes=len(class_names))\n",
    "\n",
    "fpr = dict()\n",
    "tpr = dict()\n",
    "roc_auc = dict()\n",
    "n_classes = len(class_names)\n",
    "\n",
    "for i in range(n_classes):\n",
    "    fpr[i], tpr[i], _ = roc_curve(y_true_one_hot[:, i], y_pred_probs[:, i])\n",
    "    roc_auc[i] = auc(fpr[i], tpr[i])\n",
    "\n",
    "fpr[\"micro\"], tpr[\"micro\"], _ = roc_curve(y_true_one_hot.ravel(), y_pred_probs.ravel())\n",
    "roc_auc[\"micro\"] = auc(fpr[\"micro\"], tpr[\"micro\"])\n",
    "\n",
    "plt.figure(figsize=(12, 8))\n",
    "plt.plot(fpr[\"micro\"], tpr[\"micro\"],\n",
    "         label='Micro-average ROC (area = {0:0.2f})'.format(roc_auc[\"micro\"]),\n",
    "         color='deeppink', linestyle=':', linewidth=4)\n",
    "\n",
    "colors = itertools.cycle(['aqua', 'darkorange', 'cornflowerblue', 'green', 'red', 'purple'])\n",
    "for i, color in zip(range(n_classes), colors):\n",
    "    if i < 5: \n",
    "        plt.plot(fpr[i], tpr[i], color=color, lw=2,\n",
    "                 label='ROC: {0} (area = {1:0.2f})'.format(class_names[i], roc_auc[i]))\n",
    "\n",
    "plt.plot([0, 1], [0, 1], 'k--', lw=2)\n",
    "plt.xlim([0.0, 1.0])\n",
    "plt.ylim([0.0, 1.05])\n",
    "plt.xlabel('False Positive Rate')\n",
    "plt.ylabel('True Positive Rate')\n",
    "plt.title('ROC Curve - Multiclass')\n",
    "plt.legend(loc=\"lower right\")\n",
    "plt.show()\n",
    "\n",
    "print(\"Evaluasi Lanjutan Selesai.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "ea68945c-071a-4923-984f-9d4735280cdd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Memulai eksperimen dengan Decision Tree (Baseline)...\n",
      "Memproses Data Training untuk Decision Tree (ini mungkin memakan waktu)...\n",
      "Memproses Data Test untuk Decision Tree (ini mungkin memakan waktu)...\n",
      "\n",
      "Dimensi Training: (10501, 49152)\n",
      "Dimensi Test    : (2260, 49152)\n",
      "\n",
      "Melatih model Decision Tree (Pixel-based)...\n",
      "Selesai melatih dalam 2067.51 detik.\n",
      "Mengevaluasi akurasi Decision Tree...\n",
      "----------------------------------------\n",
      "Akurasi CNN (Deep Learning)    : 90.88%\n",
      "Akurasi Decision Tree (Piksel) : 43.14%\n",
      "----------------------------------------\n",
      "Kesimpulan: CNN jauh lebih unggul karena mampu mengekstrak fitur spasial.\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 10. (EKSPERIMEN) MELATIH DECISION TREE LANGSUNG DARI PIKSEL GAMBAR\n",
    "# =============================================================================\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "from sklearn.metrics import accuracy_score\n",
    "import numpy as np\n",
    "import time\n",
    "import tensorflow as tf\n",
    "\n",
    "print(\"\\nMemulai eksperimen dengan Decision Tree (Baseline)...\")\n",
    "\n",
    "# --- 1. Fungsi Bantu untuk Meratakan (Flatten) Dataset ---\n",
    "# Scikit-Learn butuh input (N_samples, N_features), bukan (Batch, Height, Width, Channel)\n",
    "def prepare_data_for_sklearn(dataset, name=\"Dataset\"):\n",
    "    print(f\"Memproses {name} untuk Decision Tree (ini mungkin memakan waktu)...\")\n",
    "    images_flat = []\n",
    "    labels_list = []\n",
    "    \n",
    "    for images, labels in dataset:\n",
    "        # Ubah bentuk dari (Batch, 128, 128, 3) menjadi (Batch, 49152)\n",
    "        # 128 * 128 * 3 = 49152 fitur per gambar\n",
    "        flattened_imgs = tf.reshape(images, [images.shape[0], -1]).numpy()\n",
    "        images_flat.append(flattened_imgs)\n",
    "        labels_list.append(labels.numpy())\n",
    "    \n",
    "    return np.concatenate(images_flat), np.concatenate(labels_list)\n",
    "\n",
    "# --- 2. Siapkan Data Training dan Test ---\n",
    "# Kita ambil data training untuk melatih DT, dan data Test untuk evaluasi\n",
    "X_train_flat, y_train_flat = prepare_data_for_sklearn(train_dataset, \"Data Training\")\n",
    "X_test_flat, y_test_flat = prepare_data_for_sklearn(test_dataset, \"Data Test\")\n",
    "\n",
    "print(f\"\\nDimensi Training: {X_train_flat.shape}\")\n",
    "print(f\"Dimensi Test    : {X_test_flat.shape}\")\n",
    "\n",
    "# --- 3. Latih Model Decision Tree ---\n",
    "print(\"\\nMelatih model Decision Tree (Pixel-based)...\")\n",
    "start_time = time.time()\n",
    "\n",
    "dt_model = DecisionTreeClassifier(random_state=42)\n",
    "dt_model.fit(X_train_flat, y_train_flat)\n",
    "\n",
    "end_time = time.time()\n",
    "print(f\"Selesai melatih dalam {end_time - start_time:.2f} detik.\")\n",
    "\n",
    "# --- 4. Evaluasi ---\n",
    "print(\"Mengevaluasi akurasi Decision Tree...\")\n",
    "dt_acc = dt_model.score(X_test_flat, y_test_flat)\n",
    "\n",
    "print(\"-\" * 40)\n",
    "print(f\"Akurasi CNN (Deep Learning)    : {test_accuracy * 100:.2f}%\") # Diambil dari variabel Bagian #9\n",
    "print(f\"Akurasi Decision Tree (Piksel) : {dt_acc * 100:.2f}%\")\n",
    "print(\"-\" * 40)\n",
    "\n",
    "if dt_acc < test_accuracy:\n",
    "    print(\"Kesimpulan: CNN jauh lebih unggul karena mampu mengekstrak fitur spasial.\")\n",
    "else:\n",
    "    print(\"Kesimpulan: Decision Tree cukup kompetitif (jarang terjadi pada data gambar kompleks).\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "692d87a2-ef81-4ef8-b27e-b93083aa9859",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA04AAAIQCAYAAAC2Uz6yAAAAOnRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjEwLjcsIGh0dHBzOi8vbWF0cGxvdGxpYi5vcmcvTLEjVAAAAAlwSFlzAAAPYQAAD2EBqD+naQAAWahJREFUeJzt3Qm8jOX///HPse9bdim0oYTIEqJslYRKJUUqpdJCq3wjhBZbhSS0/RRlayEVkZS+QikVkrWykyP7Mv/H++p/z3fOmDn3HM5+Xs/HY5iZc83MNffc2+daPndcIBAIGAAAAAAgqmzR/wQAAAAAEAInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJyCJKlSoYFdffbWlJ02aNHE3z/r16y0uLs7eeOMNy2r0nfXdlyxZYhlJ+G+IjOPpp5926xxSxvz5893ynTJlim/Z2267ze2jkfnW35TcR3rrmP4P/bwLLrggRT4PGReBEzLVybJ3y5Mnj5177rnWvXt327p1a1pXD8nohhtucL/x448/ntZVSTe8QNm75cyZ04oXL26XXHKJPfnkk7Zx40bLCFR3bbNIP3TyqN/lnHPOifj3zz//PLjexRLYZGasv0mnQLdAgQInPP/jjz+6fZiCYO3fgPSCwAmZSv/+/e3tt9+2kSNHupPGV155xerXr2/79++3rOTMM8+0AwcO2K233mqZSXx8vH300UfuYPruu+9aIBCwzOKzzz5zt1PRoUMHt/6PHz/ennrqKatUqZKNGDHCqlSpYpMmTUq2uiKh//znP257y6zUELVmzRpbvHjxCX+bOHGi+3t68dprr9mqVavSuhoZSnpbf1esWGFNmza1/Pnz27x581KlB/HSSy91y0D/A4nJkehfgQzmyiuvtNq1a7v7d955p5122mk2bNgw++CDD9xJ5alQ8JUvXz7LCLxet8xm6tSpduzYMZswYYJdfvnltmDBAmvcuHGq12Pfvn3uoJ6ccuXKdcrvcdFFF9ktt9yS4LkNGzZYixYtrHPnzi6Aql69+il/TmaX1G09R44c7pZZnXXWWXb06FHXWFGnTp3g8wcPHrTp06dbq1at3LaZHqi3NavLyOvvzz//7PbtefPmdUFTxYoVU+Vzs2XLlimPmUh+9DghU9MOWNatWxd87v/+7/+sVq1absdcrFgxu+mmm2zTpk0JXueNbV66dKlrgdJBSEOeQql3oEaNGm5nW7VqVZs2bVqCv+/atcseeeQRq1atmhuKUKhQIRfYLV++POLY6vfee88GDhxop59+untPtbiplTfc2LFj3YmM6q+TmK+++uqEMpHmOHlDIv78809r27atu1+iRAlXRwUjoXbu3Ol6q1TnIkWKuJNu1Tv8PTWcQu+rng3VuXTp0nb77be710caQ6/vo/J6z8KFC1uXLl2S1Buo1u3mzZvbZZdd5oIAPY7F7t273bLSslVrdKTx7H7L7ffff7errrrKChYsaB07dnR/07Jv3769nXHGGZY7d24rX7689ejR44TW2y1btrjvqs9XuTJlylibNm0SDEGJNH5fQ+xWrlxpp9r7qO9z+PBhe/755xP87e+//7aHHnrI1Vv1Ovvss+25556z48ePJyinx+q5Ov/8893vXKpUKbv77rvdco00/89v2zgVsdZFjSU6oS9btqz7btpmBgwYcMK6Hm1b99aFIUOGBLc5vc/FF19s3333ne8cEW/Y1owZM9z767Wq8+zZs0/4TloP1eCj76PPefXVV2Oad6L317oZaRtSQ5G2R+/7as5fy5Yt3fAn7Tt0QqptNVZ6v8mTJydYN9T7q8/W8NlwCtjvvfdeO++889znqRFL20qkYVdaD7XdaP3RctJ20qlTJ9uxY0eCcvpsv31k+BynpPyOou3t+uuvd8cGfYZ+lw8//NCSC+tvdL/++qv7TfVZCpp0XEmM9ml9+vRxx3MdT9SY1ahRI/facOpxVzntv3Vc03H5xRdfTPAdIh0TwmnfpmWs7UGNCch60kcTA5BCdLIrOmiLDroawqQDvXqktm/fbi+//LI74Hz//ffuhN6jk38FOgqs1IqvA5znt99+sxtvvNG6devmgorXX3/dnRTooKITe1m7dq076Oh5naRorpUOKOoh+eWXX9wBMdSzzz7rWr0UyOzZs8ed5OoE/b///W+wjIZg6SCrYYg64dVnXHPNNe4gr5NfPzro6uSpbt267oA6Z84cGzp0qDvY3XPPPcEDe+vWrd2wHD1XuXJldxDX94w0v0F1UFCgkzS1Fuogrf+//fbbEw6cWu5aFoMHD7Zly5bZuHHjrGTJku5k3c9ff/3lDohvvvmme6wD1/Dhw92wzMR6a3Typd9EgeyXX37pvuvmzZstKXSA1HJr2LChW25ea+7777/vThy1nLSOaZlpffrjjz/c3zzXXXedWyb333+/O6nbtm2bW3YKjBIbhqKTR9X5VIckariqvrc+06N6a11UIK11SsHfN998Y7169XLLRyd3Hv1dwZd+5wceeMA1RGi5a5v5+uuvE7Tyx7JtnIpY66IyCip69uzp/v/iiy/cSZaGe77wwgsJ3jOxbf2dd96xvXv3us/V+qzt8tprr3XrvV/vxsKFC13QqABCJ2wvvfSSWxf0u3v7JNX7iiuucMF0v3793DaqIcdq1PCj5Txq1CibOXOmW8ahv62CGgUR2bNnd+ubeh31nk888YTbz+nEOikB7c033+xOhnVi6TVIadnoRFfbcDidnGt90jJVoKPP09Bpnehr/+dtQ//884872dVJswI59Zpqm1Wwou1IgV5S9pHRxPI7ahtt0KCBlStXzi0nnYirQUsNTepRa9eunZ0q1t/I1KCl9Uo9X9rPa3/lR8tCxxAdC7p27eq+p46R2ldrX6zGG9F+T2W0rnrHGq1vWt4PPvhgzHX8+OOPXVCt7U6jHrRtIQsKAJnA66+/rjPLwJw5cwLbt28PbNq0KTBp0qTAaaedFsibN2/gjz/+CKxfvz6QPXv2wMCBAxO89qeffgrkyJEjwfONGzd27zdmzJgTPuvMM890f5s6dWrwuT179gTKlCkTqFmzZvC5gwcPBo4dO5bgtevWrQvkzp070L9//+Bz8+bNc+9XpUqVwKFDh4LPv/jii+551U8OHz4cKFmyZKBGjRoJyo0dO9aVU51DP0fPabl4Onfu7J4L/WxRnWvVqhV8rO+lciNGjAg+p+9x+eWXn/Ce+/fvP2H5vPvuu67cggULgs/17dvXPXf77bcnKNuuXTv3G8ViyJAh7reMj493j1evXu3ec/r06RHXhe+++y6wefPmwPnnnx+oVKmS+/3Dl7n+D5XYcnviiSdOqFOk7z948OBAXFxcYMOGDe7x7t273etfeOGFRL+ffr/Q39B7LpbdtFfvxD6jTZs2rozWVRkwYEAgf/78bjmG0vfUdrJx40b3+KuvvnKvmzhxYoJys2fPPuH5WLeNaPTa++67L+rfk1KXSL/N3XffHciXL5/bNv22dW+Zav3ctWtX8PkPPvjAPf/RRx+dsH6Hf5dcuXIF1qxZE3xu+fLl7vmXX345+Fzr1q1dnf7888/gc7/99pvbJ/n99sePHw+UK1cucN111yV4/r333kuwDWob8baJpNLy0TYktWvXDtxxxx3B9Vrf78033wxuT++//36iy3/RokWu3FtvvRV8rk+fPu65adOmRfx+SdlHetur1sOT+R2bNm0aqFatWoL1Q3W45JJLAuecc47vsmL9Tdr66/1eOXPmdPuIsmXLnrA/SmwfefTo0QTrg7delipVKsGx5sEHHwwUKlTIlY8m0jEhdN3XPk317Nq16wnHdWQtDNVDptKsWTPX0qXeF7W+qaVOY/DVgqiWM/WmqNdDLZreTT0lyhgV3r2v4QJqFYxEvUWhrY/q+lfvgFrfNCzLe71aR0WtcGoVVH00dEW9LeH0WaE9J2qFFbUMekNt1HKslvzQcmpV1jCFWOn1ofQ53meIegbUEqkWPI++x3333XfCe2kITuh8By3PevXquceRvmOkz9ZyUcuhHw3L09AVtXyKfjMNvYg2XE+t1epROXLkiJsLpSFrp8LrkYv2/TXvSd9fvYE679C64JXR76WW+vDhOH70muRKgOFlrlKrrKhHTMu/aNGiCbYHbUNaX7XMvHJav9RbFFpOy17vGb7dxLJtnKyk1CX0t9F3Vjl9X/XGhA9/TGxbV+uyllG07TIxWpahLecXXnihWx7ea7Wc1eurHo3QHmgNmVQPgh/1IKinadasWa7nxqMhddrnqYdUvJ50tZhrezhZ6nXSflRDpJRBTy3u0XphQpe/PlPbub6X6hK6b1BPjubdRXqf8B5rv31kYvx+R/VIq1dHxwdvfdFN9VYPhnpS1Tt7Klh/I9P76Ptp5ERoD6MfrX/e+qBju35DjQ7QsMHQdUzrnPbPoT3uSaG5fVqO6rXTqBHvuI6siV8fmYqGrWjnqAOQhoNoB6+DnujAp5NQnXAruAq9qdteQUkonXhEGwKmA0P4QV3pz8Ubw68duYaS6fN0YNMBQZ+leUEaZhJOQ6VCeQc772RbcwYkPC2wghy/seAejUEPH0Khzwk9odfnaNhF+ORifedwOlBpqIOGhuhAr/f2JvOezHeMRr+PTrw1jEZzGrybhv3oZDBS4KU5WvpNNdRNv+Wp0PARDTcKpyErClx1wPfmjHnJKrzvr99ew0M++eQTt5w0LFTDZU41iEgq78TaCzy1PShIDt8WdLIk3vagcvouGo4VXlbvGb7dxLJtnKyk1EXDrnQyrhNVneypjJc4I3zdTGxbP9l1NtJrvdd7r1V9NR8u0rYV6blIdEKn9/Dm4Wg5KJBSQOX9DlonNcRKQ6m0H9L8Og2hPHTokCWFGqO07LQuq8FC89m89Smc6qShZd78OW//p/lMoctfw6ljvVZOcv4W4a/V/kTHBw3lDl+3+vbt68qEr+tJxfobmY4db731ljtmq3FMQU6sNHRbAZ2ObRo+qOWkoauhy0hDDbUPUjCn/biGhEaaqxWJhlJquWv70TDsjHC9K6Qs5jghU1ECAC+rXjgFMtrp6aAfaWxy+LUkQlv8TsagQYPcQVg7aU3q1cm1Wqo0Nyl88r1EGy+dnCm3k3tMtlpnNY/h0UcfdePJtQz13TTmPTm/oxJ6iCaQ6xZOrdbhLa4ax6+DsSYAa05VqGgHv/CJ157Q3sPQst7cKV1TSnPBNCdCrdIKpkK/v35zzRvTnLdPP/3UrReqk1q4a9asaamV4lcnbDoJE9VP9X/ssccilveCHZXT66L17CV1LsOpiLUuOjlXsKDvqvkWajXXiZVaofVbha+biW3rp7JdpsY2rR5ezZPTXBz1CGluk05mFVB5vGssad6h/q51UPslzW/Uc5GuoxOJGlTUWKHXaX5IYpn0NJ9PwZnWfc2xUwCgeij4irRviEVK/hZenTR/ymtsO5VgIBLW3+i0XiggU5CjfbfWU79MozouaF+rHi8dg7RsVWftW735zaLnf/jhB7fe6/ivm9ZN9YR7c2YTW+d1U2OERn1EO79A1kHghCxDBx/t8NUj4p0UniyvdTL0BHz16tXuf2+yv05UlP1Nk1VD6aCYlOEIHm+omVotvcnZ3jAYtYolV5ppfY567MJT2oZnr9JBbu7cua4VWy3LHtUvOWk5a4KzlqUOquEUlOpEJDxw0ombTnRUN520abJ3eKurfotQXq9eLH766Sf3m+vAqwOwJ9pwEK1/Dz/8sLtpGSnQ1AmoFxSmpEWLFrkTidBU5aqPWrm9HqZoVE7DcdTbF0tjQizbxsmKtS4a4qghVhpWFnpdltDsmumBTui8aySFi/RcYg0YaiBQz6uG6Wk5e0NmQ+k53ZQkR9uUEiso25gS5cRKwZnKa/iTskxGo/2fkoNoHQ8dzhu+zek3VVCf1rxee/Xg+20TJ4v11384tBqidF0p7au0biY2LE7rmH43LafQ/Y3XQxhKQZgar3RT4KljiYbdqRErsYBY30+jGnTMVYOgRjAouyCyLobqIctQK5Zao3SiH95ipsfhKbT9Mrxp7pRHJyzq3dDJsOZMiT4r/HM0xv1kx8mrpUstkmPGjHFzDDzKvhR+MnIq1NqqYEwXkvToQKNhkJFaI8O/Y2g2tuSglm0N8VJgpIxG4Te1rCvQ028STgdFtSArU5wyeoUGh6q/N4/HM3r06JjrFen7635oiltRAKoTxvATKA1x8hsqlRzpyBUMqlVWJw5qlQ092VZApVbYcFqfvFS7KqfeNQWo4VQmfN2LZds4WbHWJdJvo20mKb9valA9dZKunsjQ9VcnnWoVj5W2Aa1LCuI1BCk8PbgaOcK3Uy/jWFKH62mb04mplmViPQKR9n8a6hTeq6shULrUQeg640nNC1wrCFBvmk6mI2XdVAbWU8X66693795uVIGOlZpT5Pf54ctJGRa1XwsVfmxXMKbhfbGu/2p4035S64h66UN7s5D10OOELEMnq88884w7idaJuLr3dfKqVjwdtO+66y53kh0L9VjdcccdLuWu5q0oNanSjav736Px/xpmoRN+JQxQD4V6RmKdjxROLaGqvw4mav3SyZLqrs882feMRMtFQx7VM6IDoIagaf6EWgLFa9nTMBJvvo4CLY2z1zUukrtVVMtMB0iNfY9E6dh1sFXrpFL3hlPqXo13V3IL/d5qydSBUHNAvDHrWjfUqpiUOQxaLnqd1hkFw1oeGroUPndAvS1Kg6uTJl3TSPOltL5pfdHwlMQkNR25hvKoB0uBrk7CtH6qTvqOb7/9dvBkQRRE6XfVeqrAShPUNbdA66lacrWNqGdUQ4a0zmn4i4a7KK211kX1munkRoGiTqaTsm0kRsNhtJ6H00ltrHXR9qZeRfV4KOWz9/1T80Q8Vkrxre1GvRBqcdeJtdJTa96PvmMslMJbrebaDnQiGDpMTxRQ6aRbc2a0zirZgBpGtM4m1msUibYd1dmP1istc5XXeq+TWfW2eGmsQ9dDrW/aHjV8UOuh9jVaN9VIlJoXbFbjkBJq6Bo/So6j/arWXdVdyWbCr8EXCetv0tffcOql1H5UqcY1xD3a5Sq0jqm3Seu1jg869mid0foWmixFPaRap3Tc1BwnNSZp36/GA10PMBbaF2o0gdYPBYtK1X6qc2eRQaV1Wj8gOYSmoPajtKINGzZ0qZh1q1y5skshu2rVqohpSMMp1W2rVq0Cn376aeDCCy906cX1HqGpeEUpYx9++GGXZlVptBs0aODS8YanVI2UyjdaamwZPXp0oGLFiu5zlR5YKYfD3zNaWm1933CRUtEqpfvNN98cKFiwYKBw4cKB2267LfD111+7ckrz7lGad6UUL1KkiCvXvn37wF9//eXK6X3DP0PvG+l3U30jUQp2pdNt1KhRIDFaHl6660jrgtLHdujQwaXInTFjRvA7Ko2zUukWLVrUpfpdsWJFzMtNfvnll0CzZs0CBQoUCBQvXtylqvVS9nrvsWPHDrd+aR3R+2g51a1b16WMTu505N5N37NYsWLuc3r16hVMjR5u79697u9nn322Sz2s76DUy0r9rmUfSmnvlbZe67LWC6Vtfuyxx9zvndRtI5rQ7xB+U/r0pNRF62u9evVcGaU51t9Vr8RSDsea4j3a+h1LamotI61ToebOnevWX/0GZ511VmDcuHFu35EnT55ArHr37u0+U79luGXLlrn1/4wzznC/iS5rcPXVVweWLFni+76J7QsT24cpLXSXLl3cOqXto2XLloGVK1dG/P47d+4MdO/e3aVW1zI4/fTTXRltO9HeP7H9XKR05LH8jvL7778HOnXqFChdurRLP606aVlNmTLFd1mx/iZ9/Y22f1Xq8LZt27p66BIPkfaRShU/aNAgVyet16rDxx9/fMI6oN+uRYsWbr1XHbUdaH+vy1XEmo7co/TsOqYrNX748QxZQ5z+SevgDUD6p+EYatlTS5taF4Fwmlujlmb13uHUe36VWS255wwCqYH1F5kVc5wAnEBZuUJp+IWGNmhoj4YFAUi57U0nm8ripeFdQHrH+oushDlOAE6gjHQ6GCqNsOZMaBy50o4rxfqppmkHkJDm0miemf7X/AslMlHihWip4oH0hPUXWQmBE4ATaBKtJuhqyJUywmniuXqcunfvntZVAzIdpTl+99133UWRdc0wNViokSL8YtdAesT6i6yEOU4AAAAA4IM5TgAAAADgg8AJAAAAAHxkuTlOujCkrnCtC2F6F/IEAAAAkPUEAgF3YfCyZctatmyJ9yllucBJQVP58uXTuhoAAAAA0olNmzbZ6aefnmiZLBc4qafJWzi6Jg0AAACArCk+Pt51qngxQmKyXODkDc9T0ETgBAAAACAuhik8JIcAAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAADSc+C0YMECa926tZUtW9bi4uJsxowZvq+ZP3++XXTRRZY7d247++yz7Y033kiVugIAAADIutI0cNq3b59Vr17dRo0aFVP5devWWatWreyyyy6zH374wR566CG788477dNPP03xugIAAADIunKk5YdfeeWV7harMWPGWMWKFW3o0KHucZUqVWzhwoU2fPhwa9myZQrWFAAAAEBWlqHmOC1atMiaNWuW4DkFTHo+mkOHDll8fHyCGwAAAABk2sBpy5YtVqpUqQTP6bGCoQMHDkR8zeDBg61w4cLBW/ny5VOptgAAAAAyiwwVOJ2MXr162Z49e4K3TZs2pXWVAAAAAGQwaTrHKalKly5tW7duTfCcHhcqVMjy5s0b8TXKvqcbAAAAAGSJHqf69evb3LlzEzz3+eefu+cBAAAAIFMGTv/8849LK66bl25c9zdu3BgcZtepU6dg+W7dutnatWvtscces5UrV9ro0aPtvffesx49eqTZdwAAAACQ+aVp4LRkyRKrWbOmu0nPnj3d/T59+rjHmzdvDgZRolTkM2fOdL1Muv6T0pKPGzeOVOQAAAAAUlRcIBAIWBaiDHzKrqdEEZobBQAAACBrik9CbJCh5jgBAAAAQFogcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETkAqW7NmjXXq1MnKli1ruXLlsnLlytmdd95pf/75Z8SyHTt2tFKlSlnu3LntrLPOsscff9zi4+Nj/rwDBw7YM888YzVq1LACBQpYzpw5rXTp0ta6dWubN2/eCeW//fZbu/baa139VDZfvnxWrVo1e+qpp2zv3r3Bcvv377eHH37YzjjjDCtYsKA1aNDAFi5ceML73XXXXRYXF2fffPNNkpYTAABAehIXCAQCloXohLNw4cK2Z88eK1SoUFpXB1nM8uXL7dJLL40Y+ChQ+frrr61ChQrBso0bN3brajgFQQsWLHABi5/mzZvbnDlzIv4tW7Zs9uGHH1qrVq3cYwVSLVq0sKNHj0YsX69ePRcAKRB64IEH7OWXX7YePXpYmzZt7KabbrJ9+/bZzz//bOXLl3flv//+e6tdu7Z16NDB/u///s+3rgAAAOk1NqDHCUhF999/fzBouv3222327Nl29913u8d//fWXde/ePVi2S5cuwaBJvTYffPCBC7rkhx9+sP79+/t+3o8//hgMmtR7NGrUKPv000/t8ssvd88dP37cxowZEyyvQMgLmlRG9Rs9erR7rdcbtWzZMnf//fffd/+rJ0oBXvv27V2PlF7jUXCVN29ee+65505xyQEAAKStHGn8+UCW8c8//wSHsmmI3iuvvOL+b9q0qU2cONH9fdasWbZp0ybbvHmz662RKlWquOBGvTwXX3yxG9qnjuLx48fboEGDgkFNJKG9VdWrV7d7773X3T9y5Ih98cUX7n5o71Jo+Z49e1rLli3d/QkTJtiSJUsSlD906FDwu4iGEsrBgwfd/++++677vgMHDnR1BgAAyMjocQJSiXqavJGxCna8gCNHjhzBoEN/X7RoUYK5Qhoep6BJypQpExzKt3v3bjcsLjEKlooUKRIc+qdg7bPPPrNhw4YFy9xyyy3B+02aNAneVxmV1Wv0WqlatapddNFF7r4CPnnrrbds586d9tFHH1n27Nnde2j+02OPPWaVKlVy86AAAAAyOgInIJUowYPG0IrmAr366qsuwHjjjTdc4OFRj9P69esTvC5UyZIlg/fXrVuX6GdqrO706dNdsKVeJvU4qRdJvU1nnnmmTZ482SWf8CjYueOOO1wApDIqq9fotUpooTlQXg/XSy+95IIn/b148eK2ZcsWGzt2rEskMXjwYPvjjz9s6NChLihUL9XWrVuTYSkCAACkDQInIJUoGHnooYeCj7t162b58+d3c5lCaaibAiuP1zMV6XFouWgUaJ133nknPL9x40abMmXKCZ+lsl4vVSj1Pv33v/8NPlbvl+ZP/f333y6AU/CneVsK+oYMGWLNmjWza665xp544gmXzU+Z/EqUKGGTJk3yrTMAAEB6Q+AEpKI+ffrYk08+6RImeJTOu06dOsHHCloUUHm8uUSew4cPB++Hlotk165d1qhRI5cQQgHU0qVL3ZBBZcLTsEAleNB9T79+/Vyvk4IgJXZQWSWiUK+XepSuv/76BL1hol409WgpMBQNzVMP04gRI2zcuHEuMcQFF1zghvzpM2+99Vb75ZdfTmr5AQAApBUCJyAVKf23kiXs2LHDBTErV660tWvXuh4Zz/nnnx+cxyThQ9wUwHgqVqyY6OdNnTrVBU9y3XXXuflJSmGuXiDPjBkzgvdfe+214P3evXu7sponpes6eUGbElhEo+F906ZNc8P39D283qVnn33W9bApg6CCKtULAAAgIyGrHpAGdFFZL8mCAqj58+e7+6eddppLBqG/e5QsQj01ShChi+RqiJ0ULVrUBSeJUYDmUdY+T+iFbEOfDy/vzaeKVj7UsWPH7MEHH3TzndRzFRrkaT6VeAFhaPAHAACQEdDjBKSimTNnup4fJYTQnCElT9AFanU9JdEwuTx58rihezVr1nTPrVq1yvXU6EK1N998czAzn5I4eIkaNHxOgZVuoZnxQgOr9957z0aOHOmus6Ten9CL6UYqr2tHffLJJ+7aT941m8LLh9JQvBUrVtgzzzwTnCPlBUrbt29P8L8XSAEAAGQUcQHvLCyLSMrVgYHkpmFx7dq1i/g3XUBW1z7y5gppbpGCoNBrK4UGLwsWLHBD6bzAyRu2p4vRej1Y6gXSRXO/+eabiJ+pVOihF8T9+OOPrW3btu51kSiL3ueffx5Mj+7RnKhzzjnHBUTqQdOQxNDvq/d85JFH7LbbbnPXqFq9erWVLVs25uUGAACQ1rEBPU5AKtLFbNXjpIQQStOtDbRhw4auB0qpwb2gyQuOvvvuO9fLpCFzynin4Ei9Ul9++WUwaEqM3k+BjobOaa6ShgAqWFKyB81b0vWivKBJrr76avfeCnSUBU9l9Rq9VnOzFFiFB03y1FNPuetKvfjii8GgSfQ+SlH+66+/utTmGoqoQI2gCQAAZDT0OAEAAADIkuLpcQIAAACA5EPgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgI4dlUUePHnW3cLpGTei1dCKVSemyouvnnExZXbg0sQzz6aGsloN3LaCUKnv8+HF3S89ldb0j75pH6aGslm20C9+ml7Kh21FKlU2r7Z59xP+wj/gX+4ikl2Uf8T/sI06tLPuIrLOPOOqzfYTKsoHT9OnT3YU9w5UpU8aaNGkSfDxt2rSoP0CJEiWsWbNmwccffvihHTp0KGLZYsWKuQuAembOnGn79++PWFY55Fu1ahV8rAuGKsd8JPoObdq0CT6eM2eO7dq1K2JZXXBVFz31zJs3z7Zv3x6xrFaqG264Ifj4q6++ss2bN1s0HTp0CN5ftGiRbdq0KWrZ9u3bB3eQusDrunXropZt166d5cmTx91ftmyZrVmzJmrZa665xvLnz+/uL1++3FauXBm17FVXXeVy9svPP/9sK1asiFq2RYsW7sKtsmrVKvvhhx+iltXFZHVxWVFdly5dGrXspZdeauXKlXP3169fb//973+jlm3QoIG7aK788ccf9vXXX0ctW7duXatUqZK7r99swYIFUcvWqlXLzj33XHd/27Zt9sUXX0Qtqwvy6gK+onXss88+i1r2ggsusGrVqrn7WndnzZoVtWzlypWtZs2a7r62CW1H0Zx99tl28cUXu/va1rQdR6OLBderV8/d1zb8/vvvRy1bvnx5dyFiT2Jl2Uf8i33E/7CP+Bf7iH+xj/gX+4j/YR+RvvcR0bajSLJs4JSeLN7yXYLHh3cesTEzxgUfn/HH6ZbrSM6Irz2S46iNm/FG8PHpf5a1PIdzRyx7LNsxe33G28HH5TaXsbwH/92ZhDseF7C3Z7wbfFxmSynLf+DEQNPzzoz3gvdLby1pBfb/u+OJ5N0P37dAtn9bfEpuL2GF/ikQtezkmVPsWPZ/WzpK7DjNCu+NfmGy92ZNs6M5/201OG1nMSsa/+8OLZIpn063w7mOuPvFdhexYn8XjVp26pwZdij3YXe/yN+FrfjuYlHLTv/iQzuQ96C7Xzi+kJXY+e+OMpIZX35k+/MdcPcL7i1gpXaUiFr2wwUz7Z8C+9z9Av/kt9LbS0Yt+9HXs2zvj/+4+/n257WyW0tHLTvz29m255d/D6Z5D+SxclvKRC0767+f2t+r9rj7uQ/lsvJb/t1ZRzL74Ge26/e/3f1ch3PaGVtOj1r20/1zbOeGfw/SOY7ksApbykct+/m+udb/z0HufvZj2aziljOjlp3zzzwbuOU5dz/ueJydtaVC1LJfxM+353YMDT4+e0vFqGX37dlvQ/9+Mfi40l8VLFvg39bBcAf+Pmgv/jMq+LjipjMs+/H/tRiHOrjrkI08MCb4+MxN5S3n0ci76Iy2j6hT+t+DFAAAGVlcILE+y0x8deCdO3dGvDpwWnSxX/vh/1pkPF5g4V57PPJJWYYtGxcw+/9FdNeinHSeSlnTXcpmurJaCQIpUTa9bBuZtOy0a95jGM5JlGUYTtLLMlTvfxiqd2pl2UdknX1EfHy86xHcs2dPxNggVJbtcdIGGrqRJlYuKe95MmVDTz4i8ft7hi4b50VEyVtWJ+LuhJyylE1q2fSybWSisuH7xtCTPz8pVTb0QEvZ2MrqhCPW41xmLiuZuWx62D7TQ9n0sM1ltLJx6WD7PJmySdk+yKoHAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAANJ74DRq1CirUKGC5cmTx+rWrWuLFy9OtPyIESPsvPPOs7x581r58uWtR48edvDgwVSrLwAAAICsJ00Dp8mTJ1vPnj2tb9++tmzZMqtevbq1bNnStm3bFrH8O++8Y0888YQr/+uvv9r48ePdezz55JOpXncAAAAAWUeaBk7Dhg2zrl27WpcuXaxq1ao2ZswYy5cvn02YMCFi+W+++cYaNGhgN998s+ulatGihXXo0MG3lwoAAAAAMmTgdPjwYVu6dKk1a9bsf5XJls09XrRoUcTXXHLJJe41XqC0du1amzVrll111VVRP+fQoUMWHx+f4AYAAAAASZHD0siOHTvs2LFjVqpUqQTP6/HKlSsjvkY9TXpdw4YNLRAI2NGjR61bt26JDtUbPHiw9evXL9nrDwAAACDrSPPkEEkxf/58GzRokI0ePdrNiZo2bZrNnDnTBgwYEPU1vXr1sj179gRvmzZtStU6AwAAAMj40qzHqXjx4pY9e3bbunVrguf1uHTp0hFf89RTT9mtt95qd955p3tcrVo127dvn911113Wu3dvN9QvXO7cud0NAAAAADJcj1OuXLmsVq1aNnfu3OBzx48fd4/r168f8TX79+8/IThS8CUaugcAAAAAmarHSZSKvHPnzla7dm2rU6eOu0aTepCUZU86depk5cqVc/OUpHXr1i4TX82aNd01n9asWeN6ofS8F0ABAAAAQKYKnG688Ubbvn279enTx7Zs2WI1atSw2bNnBxNGbNy4MUEP03/+8x+Li4tz///5559WokQJFzQNHDgwDb8FAAAAgMwuLpDFxrgpHXnhwoVdoohChQpZetB6Rru0rgIApJiP2k5P6yoAAHDKsUGGyqoHAAAAAGmBwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMBHDkuiX3/91SZNmmRfffWVbdiwwfbv328lSpSwmjVrWsuWLe26666z3LlzJ/VtAQAAACDj9zgtW7bMmjVr5gKkhQsXWt26de2hhx6yAQMG2C233GKBQMB69+5tZcuWteeee84OHTqUsjUHAAAAgPTW46SepEcffdSmTJliRYoUiVpu0aJF9uKLL9rQoUPtySefTK56AgAAAED6D5xWr15tOXPm9C1Xv359dzty5Mip1g0AAAAAMtZQvViCplMpDwAAAACZJjnEjh07bMKECW5I3pYtW9xzpUuXtksuucRuu+02lygCAAAAALJsOvLvvvvOzj33XHvppZescOHCdumll7qb7uu5ypUr25IlS1KutgAAAACQ3nuc7r//fmvfvr2NGTPG4uLiEvxNWfW6devmyqg3CgAAAACyZOC0fPlye+ONN04ImkTP9ejRw6UrBwAAAIAsO1RPc5kWL14c9e/6W6lSpZKjXgAAAACQMXucHnnkEbvrrrts6dKl1rRp02CQtHXrVps7d6699tprNmTIkJSqKwAAAACk/8Dpvvvus+LFi9vw4cNt9OjRduzYMfd89uzZrVatWm4Y3w033JBSdQUAAACAjJGO/MYbb3Q3XeBWqclFwRTXbQIAAACQWSU5cPIoUCpTpkzy1gYAAAAAMnpyCD8avte/f//kfEsAAAAAyFyB09SpU908JwAAAADITE56qF4kyqwHAAAAAJlNsvY4AQAAAEBmlKyB0+7du+2tt95KzrcEAAAAgMwVOG3cuNG6dOmSnG8JAAAAABlrjlN8fHyif9+7d++p1gcAAAAAMnbgVKRIEYuLi4v690AgkOjfAQAAACDTB04FCxa03r17W926dSP+/bfffrO77747ueoGAAAAABkvcLrooovc/40bN47aI6VeJwAAAADIsskhbr75ZsuTJ0/Uv5cuXdr69u2bHPUCAAAAgIzZ49S1a9dE/16qVCkCJwAAAACZDhfABQAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJwAAAABI6cCpVatWtnnz5lN9GwAAAADIvIHTggUL7MCBA8lTGwAAAABIhxiqBwAAAAApHTideeaZljNnzlN9GwAAAADIvIHTihUrrHz58if9+lGjRlmFChUsT548VrduXVu8eHGi5f/++2+77777rEyZMpY7d24799xzbdasWSf9+QAAAADgJ4elocmTJ1vPnj1tzJgxLmgaMWKEtWzZ0latWmUlS5Y8ofzhw4etefPm7m9TpkyxcuXK2YYNG6xIkSJpUn8AAAAAWUOaBk7Dhg2zrl27WpcuXdxjBVAzZ860CRMm2BNPPHFCeT2/a9cu++abb4LDA9VbBQAAAACZMjmEeo+WLl1qzZo1+19lsmVzjxctWhTxNR9++KHVr1/fDdUrVaqUXXDBBTZo0CA7duxY1M85dOiQxcfHJ7gBAAAAQIYInHbs2OECHgVAofR4y5YtEV+zdu1aN0RPr9O8pqeeesqGDh1qzzzzTNTPGTx4sBUuXDh4O5X5WAAAAACyppMOnPbs2ePmIumm+6nh+PHjbn7T2LFjrVatWnbjjTda79693RC/aHr16uXq5902bdqUKnUFAAAAkIUDp3HjxlnVqlWtWLFi7v/Q++PHj4/5fYoXL27Zs2e3rVu3Jnhej0uXLh3xNcqkpyx6ep2nSpUqrodKQ/8iUea9QoUKJbgBAAAAQIoFTi+88II9+OCD1qZNG5s7d65LRa6b7rdt29b9bciQITG9V65cuVyvkV4b2qOkx5rHFEmDBg1szZo1rpxn9erVLqDS+wEAAABAmmfVGzlypL3++ut2ww03JHhevT5NmjSx6tWr26OPPmqPPPJITO+nVOSdO3e22rVrW506dVw68n379gWz7HXq1MmlHNc8JbnnnntcHRSg3X///fbbb7+55BAPPPBAUr4GAAAAAKRc4LRt2zarVq1a1L/rb0r6ECvNUdq+fbv16dPHDberUaOGzZ49O5gwYuPGjS7TnkeJHT799FPr0aOHXXjhhS6oUhD1+OOPJ+VrAAAAAECSxAUCgUCshS+99FKrWLGim8uUI0fCmEuZ7m6//XZbv369ffnll5ZeKR25suspUUR6me/Ueka7tK4CAKSYj9pOT+sqAABwyrFBkofqtWzZ0iVvUBDl9QwpocOCBQvcPKPPPvssKW8JAAAAAJkrOYSGxykZw4ABA6xgwYLuukq66b6upbRy5Up3UVoAAAAAyEyS1OMkCpKUpEE3AAAAAMgKTvoCuAAAAACQVSRr4NSsWTOrVKlScr4lAAAAAGS8oXqJadeuXZLSkQMAAABAlguc7rvvvuR8OwAAAADI+EP1Dh065G4AAAAAkJklOXD6/PPP7aqrrrKiRYtavnz53E339dycOXNSppYAAAAAkFECpzfffNMFSLq67vDhw+3jjz92N90vUqSI+9vbb7+dcrUFAAAAgPQ+x2ngwIE2YsSIiHOZbrvtNmvYsKH179/fbr311uSsIwAAAABknB6njRs3upTj0TRt2tT++OOP5KgXAAAAAGTMwOn888+38ePHR/37hAkTrGrVqslRLwAAAADImEP1hg4daldffbXNnj3b9TyVKlXKPb9161abO3eurV271mbOnJlSdQUAAACA9B84NWnSxFasWGGvvPKKffvtt7Zlyxb3fOnSpe3KK6+0bt26WYUKFVKqrgAAAACQMS6Aq8DoueeeS5naAAAAAEBGnuMUCARStiYAAAAAkNEDJyWGmDRpkh0+fDjRcr/99pvdc8899uyzzyZH/QAAAAAg4wzVe/nll+3xxx+3e++915o3b261a9e2smXLWp48eWz37t32yy+/2MKFC+3nn3+27t27u+AJAAAAALJU4KRrNC1ZssQFR5MnT7aJEyfahg0b7MCBA1a8eHGrWbOmderUyTp27GhFixZN2VoDAAAAQHpODtGwYUN3AwAAAICsIkkXwAUAAACArIjACQAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAMmVVS8+Pt4KFSoUvJ8YrxwAAAAAZKnASddm2rx5s5UsWdKKFClicXFxJ5QJBALu+WPHjiV3PQEAAAAg/QdOX3zxhRUrVszdnzdvXkrWCQAAAAAyZuDUuHHjiPcBAAAAILM7qeQQs2fPtoULFwYfjxo1ymrUqGE333yz7d69OznrBwAAAAAZM3B69NFHgwkifvrpJ+vZs6ddddVVtm7dOncfAAAAALLkUL1QCpCqVq3q7k+dOtVat25tgwYNsmXLlrkACgAAAAAsq/c45cqVy/bv3+/uz5kzx1q0aOHuK3mEX6pyAAAAAMgSPU4NGzZ0Q/IaNGhgixcvtsmTJ7vnV69ebaeffnpy1xEAAAAAMl6P08iRIy1Hjhw2ZcoUe+WVV6xcuXLu+U8++cSuuOKK5K4jAAAAAGS8HqczzjjDPv744xOeHz58eHLUCQAAAAAyfuAU6uDBg3b48OEEzxUqVOhU3xYAAAAAMvZQvX379ln37t2tZMmSlj9/fitatGiCGwAAAABYVg+cHnvsMfviiy/c/KbcuXPbuHHjrF+/fla2bFl76623kr+WAAAAAJDRhup99NFHLkBq0qSJdenSxRo1amRnn322nXnmmTZx4kTr2LFj8tcUAAAAADJSj9OuXbusUqVKwflMeuylKV+wYEHy1hAAAAAAMmLgpKBp3bp17n7lypXtvffeC/ZEFSlSJHlrCAAAAAAZMXDS8Lzly5e7+0888YSNGjXK8uTJYz169LBHH300uesIAACQpbRq1cri4uKCt5UrVwb/prnlLVu2tPLly1vevHndOVjFihXttttuS1DOz5dffml33HGHValSxbJlyxb8rPnz5/u+9t57701Qv9mzZyf4+9ixY9375suXz/2vOod755133GsHDx4cc52BDDfHSQGSp1mzZm4jXbp0qZvndOGFFyZn/QAAALIUzRefNWtW1L9PmjTJ5s6dm+C59evXu9vUqVPdOdm5557r+znTp0+3CRMmJLl+CxcutDFjxkT9+7Rp0+zuu+92c+HfeOMN18jetWtXK168uLVt2zaYoVnJxjSKqWfPnkmuA5AhepyOHDliTZs2td9++y34nJJCXHvttQRNAAAAp2DHjh320EMPuZ6YXLlyRSxTvXp169+/vwuS5syZYy+++GLwGpr//POPvf766zF9VqlSpez666+3oUOHxhRoyaFDh1wQFAgEXE9XJO+//777/4EHHrC6deva/fffn+B5US/Tn3/+acOGDXMZmoFM2eOUM2dO+/HHH1OmNgAAAFmYgiYFT3fddZd9+umntmHDhhPKKNAJpQbttWvXugBK9u7dG9Nn9erVK3jfm6/uZ8CAAW6kUYsWLVwQpeF+4fS8eIGfFxgdPHjQ/a958voOzZs3tzZt2sT0uUCGneN0yy232Pjx45O/NgAAAFmU5glpmJ6ui/n888/H9BoFKRqaFzq077LLLkuR+v3000+uXvnz57dXX301ajkFct4cJg3Je/fdd4PTO+Thhx+2o0eP2ogRI1KknkC6muOklV1jYtU9XKtWLbcBhVK3KwAAAGKjIXbdunVz90ePHm2FCxdOtLx6fZR0IZQyGz/++ON23XXXJXv9jh8/bnfeeaebsvHCCy9YhQoVopbV9/jll19ccKXgKXv27Na9e3c370lzszS36sEHH7SqVau68lu3brXTTjvNcuQ4qdNSINWc1Bq6YsUKu+iii9z91atXJ/ibxuQCAAAgdr1793bD8tq3b3/Sw9cUeGjuUUrQMMDFixdbvXr1gnOWolGgpIzLGo73119/uR40zYdSw7sCJiWJePrpp12SC72XhiZqOJ/+prlPyvAHZJrAad68eclfEwAAgCxIvUcjR460okWL2ssvvxzTa5SY66uvvrL9+/fbDz/8YM8++6wLQJ588kkrWLCg6+FJLrt377annnrKzXN/7bXXYg5sFCwpa57nlVdesZ9//tn1RCmguvXWW11SCz2vKSAaBnjWWWe5+V1AekRIDwAAkIa2bNnihsIpQCldunTw2kihiSE0LK9GjRrBx7p+U8OGDV2SBqX1Dk0YoeFxyWnPnj1urpKG6VWrVi1Yv9DEEFdeeaUbKhjNzp07rW/fvu47aMjflClTXA/UPffc44b2DRo0yJWbPHlystYdSPMeJ006TGxI3hdffHEqdQIAAECUZBDKVhd+Hhb6+O+//7b05j//+Y8LDD/44APXY6Vg0es5E2/OlPc8kGkCp9AWD1ELhLqJNfepc+fOyVU3AACATO/ss8+24cOHn/C8rtWkYMNLHX7++efbokWLXI+NhrldcMEFbqibst0NHDgw+DpvHnp4UKUgRRfJ9SiBg26ya9eu4PPqSdKwP9F1nooVKxaxfhpe+Pvvv7v7Svyg60tFsnz5cjfE76abbrJGjRolCJS2b9+e4H8vkALSo7hAMs4i1EQ/ZYUZMmSIpVfx8fEuU426nb2LxaW11jPapXUVACDFfNR2elpXAciQFFx4w/V+/fVXq1y5ss2fPz/RdOMa6vf1118nmFsULXDSeVu/fv0SrUNip4lNmjQJDtf75JNP7Iorroha7rvvvnNzucqXL++e08VvzzvvPFffN9980507zpgxw924thPSa2yQrHOcdH0npSkHAABA8jv33HPtgQcecL1Kyk6nDHZKBlGzZk3XK/Xjjz8mCJrSmi6sq+BKadK9oEnKlSvnLvBbokQJa9mypQsMx44dS9CErNPj9Pbbb7sNQ5lS0it6nAAgddHjBADIDLHBSc1xuvbaaxM8Vuy1efNmW7JkiUtXCQAAAACZyUkFTuFXs1Z2FI1T1SRGpcUEAAAAAMvqgZOuHh2tK2vNmjUuOwwAAAAAZBYnlRyiVatW7joC4VatWuUypwAAAACAZfXAqUCBAtauXTt3xWePsqEoaLruuuuSs34AAAAAkDEDp2nTprnMEx07dnSJIXThWwVNHTp0cMP4AAAAAMCyeuCUN29emzlzphuad8MNN1jTpk2tU6dONmzYsOSvIQAAAABklOQQynEenklv8uTJ1rx5czc8T2nIvTLp5fpIAABkWDedldY1AICUM+l3y7SBU5EiRSwuLu6E5zVUb8yYMfbqq6+6+ypz7Nix5K4nAAAAAKT/wGnevHkpWxMAAAAAyOiBU+PGjWMqp0QRAAAAAGBZPTlEuL1799rYsWOtTp06Vr169eR4SwAAAADIHIHTggULrHPnzlamTBkbMmSIXX755fbtt98mX+0AAAAAICMN1fNs2bLF3njjDRs/frzLoqd05IcOHbIZM2ZY1apVU6aWAAAAAJBRepxat25t5513nv344482YsQI++uvv+zll19OudoBAAAAQEbrcfrkk0/sgQcesHvuucfOOeeclKsVAAAAAGTUHqeFCxe6RBC1atWyunXr2siRI23Hjh2nXIlRo0ZZhQoVLE+ePO59Fy9eHNPrJk2a5K4b1bZt21OuAwAAAAAkS+BUr149e+2112zz5s129913u8ClbNmydvz4cfv8889dUJVUkydPtp49e1rfvn1t2bJlLitfy5Ytbdu2bYm+bv369fbII49Yo0aNkvyZAAAAAJDiWfXy589vt99+u+uB+umnn+zhhx+2Z5991kqWLGnXXHNNkt5r2LBh1rVrV+vSpYtLLjFmzBjLly+fTZgwIeprjh07Zh07drR+/fpZpUqVTuYrAAAAAEDqXcdJySKef/55++OPP+zdd99N0msPHz5sS5cutWbNmv2vQtmyuceLFi2K+rr+/fu7IO2OO+7w/Qxl/FP2v9AbAAAAAKT6BXAle/bsbq7Rhx9+GPNrND9KvUelSpVK8LweK+15JOrlUip0DRmMxeDBg61w4cLBW/ny5WOuHwAAAAAka+CUGjSH6tZbb3VBU/HixWN6Ta9evWzPnj3B26ZNm1K8ngAAAACy+AVwk5OCH/VUbd26NcHzely6dOkTyv/+++8uKYSuJ+VRYgrJkSOHrVq1ys4666wEr8mdO7e7AQAAAECG7HHKlSuXS20+d+7cBIGQHtevX/+E8pUrV3bJKH744YfgTckoLrvsMnefYXgAAAAAMl2PkygVeefOna127dpWp04dGzFihO3bt89l2ZNOnTpZuXLl3FwlXefpggsuSPD6IkWKuP/DnwcAAACATBM43XjjjbZ9+3br06ePSwhRo0YNmz17djBhxMaNG12mPQAAAABIK3GBQCBgWYjSkSu7nhJFFCpUyNKD1jPapXUVACDFfNR2elpXIWO6KeGcXQDIVCb9bhktNqArBwAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAAyQuA0atQoq1ChguXJk8fq1q1rixcvjlr2tddes0aNGlnRokXdrVmzZomWBwAAAIAMHzhNnjzZevbsaX379rVly5ZZ9erVrWXLlrZt27aI5efPn28dOnSwefPm2aJFi6x8+fLWokUL+/PPP1O97gAAAACyhjQPnIYNG2Zdu3a1Ll26WNWqVW3MmDGWL18+mzBhQsTyEydOtHvvvddq1KhhlStXtnHjxtnx48dt7ty5qV53AAAAAFlDmgZOhw8ftqVLl7rhdsEKZcvmHqs3KRb79++3I0eOWLFixSL+/dChQxYfH5/gBgAAAAAZJnDasWOHHTt2zEqVKpXgeT3esmVLTO/x+OOPW9myZRMEX6EGDx5shQsXDt40tA8AAAAAMtRQvVPx7LPP2qRJk2z69OkusUQkvXr1sj179gRvmzZtSvV6AgAAAMjYcqTlhxcvXtyyZ89uW7duTfC8HpcuXTrR1w4ZMsQFTnPmzLELL7wwarncuXO7GwAAAABkyB6nXLlyWa1atRIkdvASPdSvXz/q655//nkbMGCAzZ4922rXrp1KtQUAAACQVaVpj5MoFXnnzp1dAFSnTh0bMWKE7du3z2XZk06dOlm5cuXcXCV57rnnrE+fPvbOO++4az95c6EKFCjgbgAAAACQ6QKnG2+80bZv3+6CIQVBSjOuniQvYcTGjRtdpj3PK6+84rLxXX/99QneR9eBevrpp1O9/gAAAAAyvzQPnKR79+7uFu2Ct6HWr1+fSrUCAAAAgEyQVQ8AAAAAUgOBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAwAeBEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAA+CBwAgAAAAAfBE4AAAAA4IPACQAAAAB8EDgBAAAAgA8CJwAAAADwQeAEAAAAAD4InAAAAADAB4ETAAAAAPggcAIAAAAAHwROAAAAAOCDwAkAAAAAfBA4AQAAAIAPAicAAAAA8EHgBAAAAAA+CJwAAAAAICMETqNGjbIKFSpYnjx5rG7durZ48eJEy7///vtWuXJlV75atWo2a9asVKsrAAAAgKwnzQOnyZMnW8+ePa1v3762bNkyq169urVs2dK2bdsWsfw333xjHTp0sDvuuMO+//57a9u2rbutWLEi1esOAAAAIGtI88Bp2LBh1rVrV+vSpYtVrVrVxowZY/ny5bMJEyZELP/iiy/aFVdcYY8++qhVqVLFBgwYYBdddJGNHDky1esOAAAAIGvIkZYffvjwYVu6dKn16tUr+Fy2bNmsWbNmtmjRooiv0fPqoQqlHqoZM2ZELH/o0CF38+zZs8f9Hx8fb+nFkf1H0roKAJBi0tP+NkM5cjytawAAKSc+Pl0dowKBQPoOnHbs2GHHjh2zUqVKJXhej1euXBnxNVu2bIlYXs9HMnjwYOvXr98Jz5cvX/6U6g4AiE1hK5zWVQAApDfT0texYe/evVa4cOH0GzilBvVmhfZQHT9+3Hbt2mWnnXaaxcXFpWndgLRoVVGjwaZNm6xQoUJpXR0AQDrAsQFZWSAQcEFT2bJlfcumaeBUvHhxy549u23dujXB83pcunTpiK/R80kpnzt3bncLVaRIkVOuO5CR6cDIwREAEIpjA7Kqwj49TekiOUSuXLmsVq1aNnfu3AQ9Qnpcv379iK/R86Hl5fPPP49aHgAAAABOVZoP1dMwus6dO1vt2rWtTp06NmLECNu3b5/LsiedOnWycuXKublK8uCDD1rjxo1t6NCh1qpVK5s0aZItWbLExo4dm8bfBAAAAEBmleaB04033mjbt2+3Pn36uAQPNWrUsNmzZwcTQGzcuNFl2vNccskl9s4779h//vMfe/LJJ+2cc85xGfUuuOCCNPwWQMagYau6Zlr48FUAQNbFsQGITVwgltx7AAAAAJCFpfkFcAEAAAAgvSNwAgAAAAAfBE4AAAAA4IPACUhhq1atctcZ08XVcOqefvppl0QmOR0+fNgqVKjgMnQCAABEQuCETEkZGu+//36rVKmSyxKkK6K3bt06wTXAdKIcFxdn3377bYLXPvTQQ9akSZMEJ+oq161btwTlfvjhB/f8+vXrE61Lr169XF0KFizoHs+fP9+9TjdljNRF12rWrGmPPfaYbd682dIDr45///23pTePPPLICddyS45ryul9H3/88WR9XwDwjje63Epyl80MtD+vUqWKHTt2LKbyb7zxhhUpUiTZPl/HcB3vdEyXX375xU4//XR3aRwgHIETMh3tBHVh5S+++MJeeOEF++mnn1yK+8suu8zuu+++BGXz5MkT08myyo0fP95+++23JNVF6fQ//vhju+222yL2RP3111/23XffuTrMmTPHpdVXfbMi9frEokCBAnbaaacl++d37NjRFi5caD///HOyvzeA9Ef7Za8RK2fOnO4yKM2bN7cJEybY8ePHk/WztJ+/6667kr3sqX7vSDcFbqlJjYa6xEz27NmDgVFo46KCGF3bc9u2bcHL2KxevTrF6lO1alWrV6+eDRs2LMU+AxkXgRMynXvvvdftcBcvXmzXXXednXvuuXb++ee7iy2H9y7p4KTnZs2aleh7nnfeeS7w6t27d5Lq8t5771n16tXdRZzDlSxZ0g3hU/1uuukm+/rrr61EiRJ2zz33JCg3btw41xqn4K1y5co2evToBH/ftGmT3XDDDa4FrlixYtamTZsEvWA6SLZt29b69evn3r9QoUKu9yzWQCWSQ4cOuR4afa/8+fNb3bp1XS+VZ+fOndahQwf393z58lm1atXs3XffTfAe6tXr3r276+ErXry4tWzZMtjTpRZIXRRbr9W12xRkRhuq532/IUOGWJkyZVxQpQD5yJEjwTLqydMFs/PmzWsVK1Z014ILb9UtWrSoNWjQwF1UG0DWcMUVV7j9g/aZn3zyidvPP/jgg3b11Vfb0aNHk+1ztO/V/iy5y56MF1980X1n7yavv/568LECt1Cncqzwo8aq33//3R2rQ+k4pbr88ccf9tprr7nf5tZbb3V/035cx8+UpEDtlVdeSdZ1AJkDgRMylV27drneJZ0464Q+XHj3vk6iFURoOJ1fC+Ozzz5rU6dOTdI8mK+++soFALHQwUB1UQDltaxNnDjRXRx64MCB9uuvv9qgQYPsqaeesjfffNP9XcGBAg4NA9Rn6bXqkdHJQOjBToGIXq/ARAHMtGnTXCB1shTwLFq0yAUZP/74o7Vv3959ptcjd/DgQdfrN3PmTFuxYoULUHXQUzAbSt9Dw+RU7zFjxgSfV4A6dOhQt6xz5Mhht99+e6L1mTdvnjv46n+9p1osdfN06tTJ9e7p++s3HDt2bHAZh6pTp45bjgCyBg3lVgOWGnkuuugie/LJJ+2DDz5wJ+qh+xANW77zzjuDjU+XX365LV++PMF7ffTRR3bxxRe7Ri41BrVr1y74t9CGGl0+Uw1AZ5xxhvv8smXL2gMPPBCxrDdyQQ1i2rfrs9VQtnXr1hMak95++233Wg3/VmNctHm1+ru+s3fzjo3eY32HAQMGuP2mPs/r/VKQ06hRI3es0vB31Tl0OJtfg1okOoaol0/LLJQa0FQXLZsrr7zSfZZGZRw4cCDBUD0ty2bNmrnjoHdZUp0HqJdKx85YGyDDqU56ny+//DLRcsh6CJyQqaxZs8btPLVjjJWGCKxbt84FKYnRQVUHrKTMg9mwYYPb8cfKq7fXY6QruSuAuPbaa12Qp/979Ohhr776qvv75MmTXcCng4J6dXRgUMuhDrShBywFJxp+op439bz079/fXnrppZMajqL31me8//777iB61llnuYNlw4YN3fOiA6ee08Fc88w0x0uBlXrgQp1zzjn2/PPPux493TwKFBs3buyGTDzxxBP2zTffuGAsGvUWjRw50i0/tRTrO3rzoFauXOkOuGq11IFcv6OWlw7A4fRb6TcDkHUpKNJIATUwedQ4pMYWBVRLly51+5GmTZu6k2tRI5ECpauuusq+//57t/9RQ0wkarwZPny424+rsWnGjBlu/x2J9tEKmryT+M8//9zWrl3rhquFUsOR3kdDw3VTWTX2nSz14GsZ6LuosU7vr324eobUWKZjjwIpNaLF2qB2Ko2LCta0LMJ7gBRgqbFMvWQ6pokaIHUM8gInvwbISHTM1PGLhjSEy3HCM0AG5rU4JYVaEHWSrx1r+MEo3DPPPOOCk88++yymoQI6OQ9vSYul/joYqCVPB6s77rjDunbtGiyjA4daDEUtngoWvcQTHgUZeq1HB8DQoR/169e3f/75xw3zO/PMMy0pNAdLk3g1xDCUWhu9uUf6uw5OCpT+/PNP1/ulv4cPP1GvVCQXXnhh8L6G34lOWtRCG4kCQm98vPcab66Yhvmp10onOp6zzz7bBVuRDs779++PaTkAyLzUCKOTf1GAoN5y7YPUQ+QFFgpUpkyZ4npkdFKuXp7Qnnztd6M1Pqk3RT0lmlul/Vq0IEsBmPZlatxTL4+89dZbbp+nYEG9Q6KgQj0x3rFAPfx6rep1ssHjww8/HHys3jbNA9XQaq/RS4GKGrg0pE3Lxmu08xoLdVzVCBA9r+PByTYuKvDSiAQFWOHHOlGQpCBUPWRKDKWh9wr4tN8Pb4AUNUIqAYRe07lz56ifS0MaIiFwQqainbmCDvUyJIXmP6nr3q/7Xr0rCmLUC6JkEX40XGP37t0x10OtYaLhFgpsxOspCeUFCSqj4CNSb5kCwpSgz9Tnq9U1NFgRDSURJeXQOHoNN1FLqoZt6IAbPlY+0nBK0cmER7+nJNY7Flree83J9KapVTellhuAjEONWN6+Rw1U2u+FJ6VRw5jXQKWMbKENXIlRT4z2jeqNV4+MeqmU9dU70Q8/Jihg8oImUU+8hqrpb17gpGNGaFChxqNIw5FjFd4LpGWgQDL0WKNlpP2sgjr1gvk1qCWlcXHPnj3ueKL3V0OgRjRopEBiy3T69Omul02BnM4FJJYGyGhoSEMkBE7IVJQcQWOdR40a5cZEh5+Ya5x6pDSm2kGr615jxa+55ppEP0M9UwqgYkkioDTjatmKhQ4gmntz6aWXBk/e1eKlA5Ja+iJRL4qGTKj3S2PRo9FBT++vA4EoIYa+c+jBOFb6TjpA6qCsoXqRaM6Shpfccsst7rEOfsqCpAN+atMQQB0k1QLp9XCply5SQKv5WPp+ALI2BSXqmRAFTQpEIs3X8Y4n3r41FtrvqidcQ4g19E4JjdTYpOF14Y1AsUquxiNP+LFTy+Duu+9OMBfLox4zBVV+DWpJaVxUELhs2TKXVU/L3m/5KsDxPjt0aGAsDZCJNaTpWA+EYo4TMh0FTTqx19AHjSXXTlQHQQ0r0BC1aDTcQi1QyriWGKWsVQ+VN546MQriNOY70vUpFHhoWIHqpyBMGd127NjhWss8GvYxePBg91kKPDRkQ8MevDSpCqh04FGQorHYavnTwV0HN2Uj8qinRy1uCuI0jEFDFzQeXQelxOjz1JLq3RSAqUVRn6thEZoDoM/UMBbVU+P8Ra19OiHQ3CQtex1wQyczp/aQGw2J0e+reiqA0n0diL0WZY+WYYsWLdKkngDSB13KQvs+L9ObGqi0r1aPkIb5ht60//WGFyfl+nLa/6iXSft27bN1nIh0KQoNDdeQat082o+rETA1G6K0DPS54d9fN80HCm1QC/+7l4AiKY2LOjbpteqViyUo1bBCvUZz0LRM9Rt6x2uvATK8Xl5gHA0NaYiEHidkOtrRqqVKY7u1M1VKU/XgqLchNCiJ1GKnTEI333yz72do7LbeK7GEBaJsQDrYqmVRQVR4T4hO3NUapzrrhF0BWehBRuPKNS9IrZGPPvqoawXU0DdvnLn+tmDBApewQuO3lUVJ4701aTm0B0qPFcyoN0tDJ5QqXL1rflQ+vIVOvTcK3jTfS8tXc5h08qDrXigxg5dwQwcqfWfVUYGKUoZr+EVa0JwABY76Plq+CvJ0vabQISI6cVH9rr/++jSpI4DUp/2hgiKd9KtxR3NytH/QvkyNQ6KGFzW6aR+mZDZqPFKWTi8hhIa1qTFK+1n1UGiuk/aTaqSKlExIc5H0eeoB0f7x//7v/1xwEGm+qT5b+3w1Vml4n95XPVSaWxRrxtbkoO+hfbwa3HRc0rFIAY8ayJSYJ7RBTfOJFHBs377dBZMKKpWwJxIdIxJL0hAL/Q5KfqR9uAI8HSs1d0m9YJrLqgZINSaqYVRDI/WbK2Orerp0zI1ECZp0bNPyBxIIAEhRI0eODLRo0SLNPr9z586BNm3apNnnp0ebNm1SFo7AnDlzgs/dcMMNgYEDB6ZpvQCk7r5R+wHdcuTIEShRokSgWbNmgQkTJgSOHTuWoGx8fHzg/vvvD5QtWzaQM2fOQPny5QMdO3YMbNy4MVhm6tSpgRo1agRy5coVKF68eODaa68N/u3MM88MDB8+3N2fPn16oG7duoFChQoF8ufPH6hXr16CfVFoWdmwYUPgmmuucWULFiwYaN++fWDLli3Bv/ft2zdQvXr1BPXV6/U+sdD3V52ifb5n8eLFgebNmwcKFCjg6nLhhRcm2GcePnw40KdPn0CFChXcMipTpkygXbt2gR9//DHqZ+/cuTOQJ0+ewMqVK4PPvf7664HChQtHfU3o37dt2xYoVapUYNCgQQnqUatWLbdP90ycODH42xQtWjRw6aWXBqZNm+b+tm7dOrcMvv/++2B5vV/Lli19lx2ynjj9kzCUApCc1EL43HPPuRavSBmBUpouEKthHcoAlVVp2IbGuqvlVj2QulK9WhM1/FE9jRrKqJZk9aAlZa4CAODUqIcoPj4+eJmNtKbjgUZoaNi+htADoZjjBKQwDdXTBV3TImiCBS8UrAtbKoWvhtZo6KbmFXgTqjVGX8MLCZoAIHXp+KhhiqeSzCI5KaW6jhcETYiEHicAAAAA8EGPEwAAAAD4IHACAAAAAB8ETgAAAADgg8AJAAAAAHwQOAEAAACADwInAAAAAPBB4AQAAAAAPgicAAAAAMAHgRMAAAAAWOL+Hzm8Xb6aF8UDAAAAAElFTkSuQmCC",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Selisih Akurasi: 47.74%\n",
      "ANALISIS: Model CNN jauh lebih efektif karena menggunakan Convolutional Layers\n",
      "          untuk mengekstrak fitur bentuk/tekstur, sedangkan Decision Tree\n",
      "          hanya melihat nilai piksel mentah tanpa memahami konteks spasial.\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 11. PERBANDINGAN AKURASI MODEL (CNN vs Decision Tree)\n",
    "# =============================================================================\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 1. Siapkan Data untuk Grafik\n",
    "# Pastikan variabel 'test_accuracy' (dari Bagian #9) dan 'dt_acc' (dari Bagian #10) sudah ada\n",
    "model_names = ['CNN (Deep Learning)', 'Decision Tree (Pixel)']\n",
    "accuracies = [test_accuracy, dt_acc] # Menggunakan hasil evaluasi Test Data\n",
    "\n",
    "# Definisi Warna: Biru untuk CNN, Merah Muda untuk DT\n",
    "colors = ['#4CAF50', '#FF5722'] # Hijau & Oranye\n",
    "\n",
    "# 2. Buat Bar Chart\n",
    "plt.figure(figsize=(10, 6))\n",
    "bars = plt.bar(model_names, accuracies, color=colors)\n",
    "\n",
    "# 3. Kustomisasi Grafik\n",
    "plt.ylabel('Akurasi (0.0 - 1.0)')\n",
    "plt.title('Perbandingan Akurasi: Deep Learning vs Machine Learning Klasik')\n",
    "plt.ylim(0, 1.1) # Memberi ruang di atas bar untuk teks\n",
    "\n",
    "# Tambahkan garis batas akurasi CNN agar perbedaannya terlihat jelas\n",
    "plt.axhline(y=test_accuracy, color='gray', linestyle='--', alpha=0.7)\n",
    "\n",
    "# 4. Tambahkan Label Angka di Atas Bar\n",
    "for bar in bars:\n",
    "    height = bar.get_height()\n",
    "    plt.text(bar.get_x() + bar.get_width()/2.0, height + 0.01, \n",
    "             f'{height*100:.2f}%', \n",
    "             ha='center', va='bottom', fontsize=12, fontweight='bold')\n",
    "\n",
    "# 5. Tampilkan Grafik\n",
    "plt.show()\n",
    "\n",
    "# 6. Cetak Kesimpulan Teks\n",
    "print(f\"Selisih Akurasi: {(test_accuracy - dt_acc) * 100:.2f}%\")\n",
    "if test_accuracy > dt_acc:\n",
    "    print(\"ANALISIS: Model CNN jauh lebih efektif karena menggunakan Convolutional Layers\")\n",
    "    print(\"          untuk mengekstrak fitur bentuk/tekstur, sedangkan Decision Tree\")\n",
    "    print(\"          hanya melihat nilai piksel mentah tanpa memahami konteks spasial.\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "25567f98-c505-407c-a930-4b86f077667f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- Perbandingan Hasil Akurasi Final Kedua Model ---\n",
      "1. Akurasi Model CNN (Deep Learning)    : 90.88%\n",
      "2. Akurasi Model Decision Tree (Piksel) : 43.14%\n",
      "----------------------------------------------------\n",
      "\n",
      "KESIMPULAN: Model CNN lebih unggul sebesar 47.74% poin.\n",
      "            Ini membuktikan Deep Learning lebih baik untuk data gambar.\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 12. PERBANDINGAN HASIL AKURASI FINAL (VERSI PERBAIKAN)\n",
    "# =============================================================================\n",
    "\n",
    "# Pastikan Bagian #9 dan Bagian #10 sudah dijalankan sebelumnya!\n",
    "\n",
    "print(\"--- Perbandingan Hasil Akurasi Final Kedua Model ---\")\n",
    "\n",
    "# Menggunakan variabel 'test_accuracy' dari Bagian #9\n",
    "print(f\"1. Akurasi Model CNN (Deep Learning)    : {test_accuracy * 100:.2f}%\")\n",
    "\n",
    "# Menggunakan variabel 'dt_acc' dari Bagian #10\n",
    "print(f\"2. Akurasi Model Decision Tree (Piksel) : {dt_acc * 100:.2f}%\")\n",
    "\n",
    "print(\"----------------------------------------------------\")\n",
    "\n",
    "# Menambahkan kesimpulan otomatis\n",
    "selisih = (test_accuracy - dt_acc) * 100\n",
    "if test_accuracy > dt_acc:\n",
    "    print(f\"\\nKESIMPULAN: Model CNN lebih unggul sebesar {selisih:.2f}% poin.\")\n",
    "    print(\"            Ini membuktikan Deep Learning lebih baik untuk data gambar.\")\n",
    "else:\n",
    "    print(f\"\\nKESIMPULAN: Decision Tree kompetitif (Selisih: {selisih:.2f}%).\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "c1994aaa-983c-4953-bd52-7daedf6f0a82",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 13. VISUALISASI PERBANDINGAN AKURASI (VERSI FINAL & BENAR)\n",
    "# =============================================================================\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 1. Siapkan nama model dan nilai akurasinya\n",
    "# PERBAIKAN DI SINI: Kita menggunakan variabel 'test_accuracy' dan 'dt_acc'\n",
    "# yang dihasilkan dari Bagian #9 dan #10.\n",
    "model_names = ['CNN (Deep Learning)', 'Decision Tree (Piksel)']\n",
    "accuracies = [test_accuracy, dt_acc] \n",
    "\n",
    "# Definisi warna agar grafik lebih menarik (Biru vs Merah)\n",
    "colors = ['#2196F3', '#FF5722'] \n",
    "\n",
    "# 2. Buat bar chart\n",
    "plt.figure(figsize=(10, 6))\n",
    "bars = plt.bar(model_names, accuracies, color=colors)\n",
    "\n",
    "# 3. Tambahkan Judul dan Label\n",
    "plt.ylabel('Akurasi (0.0 - 1.0)')\n",
    "plt.title('Perbandingan Akurasi Final: CNN vs Decision Tree')\n",
    "plt.ylim(0, 1.15) # Memberi ruang lebih di atas bar untuk teks persentase\n",
    "\n",
    "# 4. Tambahkan teks persentase di atas setiap bar\n",
    "for bar in bars:\n",
    "    height = bar.get_height()\n",
    "    plt.text(bar.get_x() + bar.get_width()/2.0, height + 0.01, \n",
    "             f'{height*100:.2f}%', \n",
    "             ha='center', va='bottom', fontsize=12, fontweight='bold')\n",
    "\n",
    "# 5. Tambahkan garis batas performa CNN\n",
    "plt.axhline(y=test_accuracy, color='gray', linestyle='--', alpha=0.5)\n",
    "\n",
    "# 6. Tampilkan chart\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "bc244012-f1f9-48f8-a339-bf839162322f",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "[1] Menyimpan model ke: ../models\\model_cnn.keras...\n",
      "✅ Model berhasil disimpan.\n",
      "\n",
      "[2] Menyimpan label kelas ke: ../models\\class_names.json...\n",
      "✅ Label kelas berhasil disimpan (16 kelas).\n",
      "   Daftar: ['.ipynb_checkpoints', 'Bean', 'Bitter_Gourd', 'Bottle_Gourd', 'Brinjal', 'Broccoli', 'Cabbage', 'Capsicum', 'Carrot', 'Cauliflower', 'Cucumber', 'Papaya', 'Potato', 'Pumpkin', 'Radish', 'Tomato']\n",
      "\n",
      "Siap untuk deployment!\n"
     ]
    }
   ],
   "source": [
    "# =============================================================================\n",
    "# 14. SIMPAN MODEL DAN LABEL KELAS (VERSI FINAL ANTI-ERROR)\n",
    "# =============================================================================\n",
    "import os\n",
    "import json\n",
    "\n",
    "# 1. Definisikan direktori penyimpanan\n",
    "MODELS_DIR = '../models'\n",
    "MODEL_NAME = 'model_cnn'\n",
    "\n",
    "# Pastikan direktori 'models' ada\n",
    "os.makedirs(MODELS_DIR, exist_ok=True) \n",
    "\n",
    "# --- Opsi 1: Simpan Model Format Keras (.keras) ---\n",
    "MODEL_KERAS_PATH = os.path.join(MODELS_DIR, f\"{MODEL_NAME}.keras\")\n",
    "\n",
    "print(f\"\\n[1] Menyimpan model ke: {MODEL_KERAS_PATH}...\")\n",
    "model.save(MODEL_KERAS_PATH)\n",
    "print(\"✅ Model berhasil disimpan.\")\n",
    "\n",
    "# --- Opsi 2: Simpan Nama Kelas (SOLUSI PERBAIKAN) ---\n",
    "CLASS_NAMES_PATH = os.path.join(MODELS_DIR, \"class_names.json\")\n",
    "print(f\"\\n[2] Menyimpan label kelas ke: {CLASS_NAMES_PATH}...\")\n",
    "\n",
    "# PERBAIKAN: Alih-alih mengambil dari train_dataset (yang error),\n",
    "# Kita ambil langsung dari nama folder di direktori dataset.\n",
    "try:\n",
    "    # Pastikan DATASET_DIR mengarah ke folder dataset utama Anda\n",
    "    # (Variabel ini seharusnya sudah ada dari Bagian #2 atau #3)\n",
    "    class_names = sorted([item for item in os.listdir(DATASET_DIR) \n",
    "                          if os.path.isdir(os.path.join(DATASET_DIR, item))])\n",
    "    \n",
    "    # Simpan ke JSON\n",
    "    with open(CLASS_NAMES_PATH, 'w') as f:\n",
    "        json.dump(class_names, f)\n",
    "        \n",
    "    print(f\"✅ Label kelas berhasil disimpan ({len(class_names)} kelas).\")\n",
    "    print(f\"   Daftar: {class_names}\")\n",
    "\n",
    "except Exception as e:\n",
    "    print(f\"⚠️ Gagal menyimpan class_names: {e}\")\n",
    "    print(\"   Pastikan variabel DATASET_DIR masih terdefinisi dengan benar.\")\n",
    "\n",
    "print(\"\\nSiap untuk deployment!\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "82531431-1eee-4a09-805b-ee89d66e60ba",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}