{"nbformat":4,"nbformat_minor":0,"metadata":{"colab":{"provenance":[],"collapsed_sections":["R7iyLJUIlEt3"],"gpuType":"T4","mount_file_id":"1pHCcckeJvIzX9boDR4XUka4_uNo-2Cxc","authorship_tag":"ABX9TyO9NQ2eyIr7CJx89phP3MpJ"},"kernelspec":{"name":"python3","display_name":"Python 3"},"language_info":{"name":"python"},"accelerator":"GPU"},"cells":[{"cell_type":"markdown","source":["# **Import dan Pelabelan data**"],"metadata":{"id":"R7iyLJUIlEt3"}},{"cell_type":"code","source":["# Import Library yang diperlukan\n","import cv2\n","import numpy as np\n","import shutil\n","import os\n","import random\n","import matplotlib.pyplot as plt\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 matplotlib.image as mpimg\n","from google.colab.patches import cv2_imshow\n","from PIL import Image, ImageEnhance"],"metadata":{"id":"ozt9vuMpqA_V"},"execution_count":null,"outputs":[]},{"cell_type":"code","execution_count":null,"metadata":{"id":"36-0c9eihbVX"},"outputs":[],"source":["# from google.colab import drive\n","# drive.mount('/content/drive')"]},{"cell_type":"code","source":["import zipfile\n","import os\n","\n","# Folder kerja Google Colab\n","working_directory = '/content'\n","\n","# Cari file zip di folder kerja\n","zip_files = [f for f in os.listdir(working_directory) if f.endswith('.zip')]\n","\n","if zip_files:\n","    for zip_file in zip_files:\n","        zip_file_path = os.path.join(working_directory, zip_file)\n","        extract_to_path = os.path.join(working_directory, os.path.splitext(zip_file)[0])  # Folder sesuai nama file zip\n","\n","        # Membuka dan mengekstrak file zip\n","        with zipfile.ZipFile(zip_file_path, 'r') as zip_ref:\n","            zip_ref.extractall(extract_to_path)\n","    print(\"Selesai mengekstrak semua file zip.\")\n","else:\n","    print(\"Tidak ada file zip yang ditemukan di folder kerja.\")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"rvSshq-N_45I","executionInfo":{"status":"ok","timestamp":1736929226246,"user_tz":-420,"elapsed":500,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"ab0e8c76-2fad-4c00-ff50-2b11c6b0aa28"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Selesai mengekstrak semua file zip.\n"]}]},{"cell_type":"code","source":["# ambil dataset\n","dataset_path = \"/content/Deep Learning/Deep Learning/Dataset mentah\"\n","\n","anorganik_count = 0\n","organik_count = 0\n","\n","for root, dirs, files in os.walk(dataset_path):\n","  if \"Anorganik\" in root:\n","    anorganik_count += len(files)\n","  elif \"Organik\" in root:\n","    organik_count += len(files)\n","\n","\n","print(f\"Jumlah data anorganik: {anorganik_count}\")\n","print(f\"Jumlah data organik: {organik_count}\")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"uBcqTU1Rh1Pm","executionInfo":{"status":"ok","timestamp":1736929231504,"user_tz":-420,"elapsed":358,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"1b990cc3-2a03-4172-fdd7-db55503b03bd"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Jumlah data anorganik: 500\n","Jumlah data organik: 489\n"]}]},{"cell_type":"code","source":["import os\n","import numpy as np\n","from tensorflow.keras.preprocessing.image import ImageDataGenerator\n","from tensorflow.keras.preprocessing import image\n","\n","# 1. Set path ke dataset\n","base_dir = '/content/Deep Learning/Deep Learning/Dataset mentah'\n","organik_dir = os.path.join(base_dir, 'Organik')\n","anorganik_dir = os.path.join(base_dir, 'Anorganik')\n","output_dir = '/content/dataset_sementara'\n","\n","# 2. Buat folder output jika belum ada\n","os.makedirs(os.path.join(output_dir, 'Organik'), exist_ok=True)\n","os.makedirs(os.path.join(output_dir, 'Anorganik'), exist_ok=True)\n","\n","# 3. Hitung jumlah gambar di setiap folder\n","num_organik = len(os.listdir(organik_dir))\n","num_anorganik = len(os.listdir(anorganik_dir))\n","\n","# 4. Inisialisasi ImageDataGenerator untuk augmentasi\n","datagen = ImageDataGenerator(\n","    rotation_range=20,\n","    width_shift_range=0.2,\n","    height_shift_range=0.2,\n","    shear_range=0.2,\n","    zoom_range=0.2,\n","    horizontal_flip=True,\n","    fill_mode='nearest'\n",")\n","\n","# 5. Salin gambar anorganik ke folder output\n","for img_name in os.listdir(anorganik_dir):\n","    img_src_path = os.path.join(anorganik_dir, img_name)\n","    img_dest_path = os.path.join(output_dir, 'Anorganik', img_name)\n","    os.system(f'cp \"{img_src_path}\" \"{img_dest_path}\"')\n","\n","# 6. Salin gambar organik asli ke folder output\n","for img_name in os.listdir(organik_dir):\n","    img_src_path = os.path.join(organik_dir, img_name)\n","    img_dest_path = os.path.join(output_dir, 'Organik', img_name)\n","    os.system (f'cp \"{img_src_path}\" \"{img_dest_path}\"')\n","\n","\n","# Target jumlah gambar organik\n","target_organik_count = 500\n","needed_samples = target_organik_count - num_organik\n","\n","print(f\"Jumlah gambar yang diperlukan untuk organik: {needed_samples}\")\n","\n","# Cek jika tidak perlu augmentasi\n","if needed_samples > 0:\n","    generated_count = 0  # Untuk melacak jumlah gambar yang dihasilkan\n","    for img_name in os.listdir(organik_dir):\n","        img_path = os.path.join(organik_dir, img_name)\n","        img = image.load_img(img_path)\n","        img_array = image.img_to_array(img)\n","        img_array = np.expand_dims(img_array, axis=0)\n","\n","        # Generate augmentasi hanya sampai mencapai needed_samples\n","        for batch in datagen.flow(img_array, batch_size=1, save_to_dir=os.path.join(output_dir, 'Organik'), save_prefix='aug', save_format='jpeg'):\n","            generated_count += 1\n","            if generated_count >= needed_samples:  # Hentikan setelah mencapai target\n","                break\n","        if generated_count >= needed_samples:\n","            break\n","\n","print(\"Oversampling selesai. Gambar baru telah disimpan di folder 'dataset'.\")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"kSJSzmaK3jtf","executionInfo":{"status":"ok","timestamp":1736929238481,"user_tz":-420,"elapsed":3308,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"91073ba7-bbe3-4f1e-f7ae-685f440a00ad"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Jumlah gambar yang diperlukan untuk organik: 11\n","Oversampling selesai. Gambar baru telah disimpan di folder 'dataset'.\n"]}]},{"cell_type":"code","source":["# 1. Set path ke folder dataset\n","output_dir = '/content/dataset_sementara'\n","organik_output_dir = os.path.join(output_dir, 'Organik')\n","anorganik_output_dir = os.path.join(output_dir, 'Anorganik')\n","\n","# 2. Hitung jumlah gambar di setiap folder\n","num_organik_output = len(os.listdir(organik_output_dir))\n","num_anorganik_output = len(os.listdir(anorganik_output_dir))\n","\n","# 3. Cetak jumlah gambar\n","print(f\"Jumlah gambar di folder 'dataset/Organik': {num_organik_output}\")\n","print(f\"Jumlah gambar di folder 'dataset/Anorganik': {num_anorganik_output}\")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"5CSxVlhZ4H2H","executionInfo":{"status":"ok","timestamp":1736929257240,"user_tz":-420,"elapsed":351,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"b10a3fdb-619a-474f-faa1-30c6c8590f6c"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Jumlah gambar di folder 'dataset/Organik': 500\n","Jumlah gambar di folder 'dataset/Anorganik': 500\n"]}]},{"cell_type":"code","source":["# Memberi label pada dataset\n","\n","dataset_path = \"/content/dataset_sementara\"\n","output_dataset_path = \"/content/dataset\"\n","\n","# Create the output directory if it doesn't exist\n","if not os.path.exists(output_dataset_path):\n","  os.makedirs(output_dataset_path)\n","\n","anorganik_count = 0\n","organik_count = 0\n","\n","for root, dirs, files in os.walk(dataset_path):\n","  for file in files:\n","    file_path = os.path.join(root, file)\n","    if \"Anorganik\" in root:\n","      new_file_name = f\"anorganik_{anorganik_count}.jpg\"  # You can adjust the file extension if needed\n","      new_file_path = os.path.join(output_dataset_path, new_file_name)\n","      shutil.copy(file_path, new_file_path)\n","      anorganik_count += 1\n","    elif \"Organik\" in root:\n","      new_file_name = f\"organik_{organik_count}.jpg\"  # You can adjust the file extension if needed\n","      new_file_path = os.path.join(output_dataset_path, new_file_name)\n","      shutil.copy(file_path, new_file_path)\n","      organik_count += 1\n","\n","print(f\"Jumlah data anorganik yang telah diberi label: {anorganik_count}\")\n","print(f\"Jumlah data organik yang telah diberi label: {organik_count}\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"mHIH9eeUitg1","executionInfo":{"status":"ok","timestamp":1736929263981,"user_tz":-420,"elapsed":351,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"ea68aff8-0c70-4f4b-bd4c-1874d0f4252d"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Jumlah data anorganik yang telah diberi label: 500\n","Jumlah data organik yang telah diberi label: 500\n"]}]},{"cell_type":"code","source":["# menampilkan kondisi dataset dalam diagram batang\n","\n","# Data jumlah data\n","labels = ['Anorganik', 'Organik']\n","counts = [anorganik_count, organik_count]\n","\n","# Membuat diagram batang\n","plt.bar(labels, counts)\n","\n","# Menambahkan judul dan label sumbu\n","plt.title('Jumlah Data Sampah Setelah Diberi Label')\n","plt.xlabel('Kategori Sampah')\n","plt.ylabel('Jumlah Data')\n","\n","# Menampilkan diagram\n","plt.show()\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":472},"id":"oNHtl0rASwe_","executionInfo":{"status":"ok","timestamp":1736929267138,"user_tz":-420,"elapsed":518,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"f1fc0c74-856f-4fb5-987d-58330ce258ea"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 640x480 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":["# import shutil\n","\n","# # Correct the folder path by removing the extra space at the beginning\n","# # and specifying the correct path if it's not in the current directory\n","# folder_path = \"/content/dataset_sementara\"\n","\n","# # Assuming the folder is located at '/content/Deep Learning'\n","# shutil.rmtree(folder_path, ignore_errors=True)"],"metadata":{"id":"wDDQtdjo1u94"},"execution_count":null,"outputs":[]},{"cell_type":"code","source":["import os\n","import shutil\n","import random\n","\n","# Path ke dataset yang sudah dilabeli\n","base_dir = output_dataset_path  # Path yang benar ke dataset Anda (organik dan anorganik berada di sini)\n","\n","# Path untuk folder \"Deep Learning\"\n","deep_learning_dir = \"/content\"  # Lokasi folder \"Deep Learning\"\n","model_dataset_dir = os.path.join(deep_learning_dir, 'dataset_model')  # Folder baru dataset_model\n","\n","# Path untuk training dan validation di dalam dataset_model\n","train_dir = os.path.join(model_dataset_dir, 'training')\n","val_dir = os.path.join(model_dataset_dir, 'validation')\n","\n","# Buat folder dataset_model beserta subfolder training dan validation\n","os.makedirs(train_dir, exist_ok=True)\n","os.makedirs(val_dir, exist_ok=True)\n","\n","# Set parameter\n","IMAGE_SIZE = (200, 200)\n","BATCH_SIZE = 32\n","SEED = 999\n","\n","# Pisahkan data menjadi set pelatihan (80%) dan validasi (20%) dari folder output_dataset_path\n","images = os.listdir(base_dir)\n","random.shuffle(images)\n","\n","# Filter gambar berdasarkan label di nama file dan pisahkan ke folder training dan validation\n","for label in ['anorganik', 'organik']:\n","    # Buat folder untuk masing-masing kelas di dalam training dan validation jika belum ada\n","    os.makedirs(os.path.join(train_dir, label), exist_ok=True)\n","    os.makedirs(os.path.join(val_dir, label), exist_ok=True)\n","\n","    # Ambil hanya gambar yang sesuai dengan label\n","    label_images = [img for img in images if label in img]\n","    split_index = int(0.8 * len(label_images))  # 80% untuk training\n","\n","    train_images = label_images[:split_index]\n","    val_images = label_images[split_index:]\n","\n","    # Pindahkan gambar ke folder training di dataset_model\n","    for image in train_images:\n","        src = os.path.join(base_dir, image)\n","        dst = os.path.join(train_dir, label, image)\n","        shutil.move(src, dst)\n","\n","    # Pindahkan gambar ke folder validation di dataset_model\n","    for image in val_images:\n","        src = os.path.join(base_dir, image)\n","        dst = os.path.join(val_dir, label, image)\n","        shutil.move(src, dst)\n","\n","# Pastikan struktur foldernya benar\n","print(f\"Gambar telah dipindahkan ke folder dataset_model:\\nTraining: {train_dir}\\nValidation: {val_dir}\")\n","print(f\"Isi folder training: {os.listdir(train_dir)}\")\n","print(f\"Isi folder validation: {os.listdir(val_dir)}\")\n","\n","# Preprocessing menggunakan ImageDataGenerator\n","from tensorflow.keras.preprocessing.image import ImageDataGenerator\n","\n","datagen = ImageDataGenerator(\n","    rescale=1./255  # Normalisasi nilai piksel ke [0, 1]\n",")\n","\n","# Persiapkan data training dan validation dari folder dataset_model\n","train_data = datagen.flow_from_directory(\n","    train_dir,\n","    class_mode='binary',  # Ubah ke binary jika menggunakan binary_crossentropy\n","    target_size=IMAGE_SIZE,\n","    batch_size=BATCH_SIZE,\n","    seed=SEED\n",")\n","\n","valid_data = datagen.flow_from_directory(\n","    val_dir,\n","    class_mode='binary',  # Ubah ke binary jika menggunakan binary_crossentropy\n","    target_size=IMAGE_SIZE,\n","    batch_size=BATCH_SIZE,\n","    seed=SEED\n",")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"fes8Kd8DUJw0","executionInfo":{"status":"ok","timestamp":1736929273917,"user_tz":-420,"elapsed":345,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"fac8cb74-fc46-4bde-c8ab-fc1e2c885007"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Gambar telah dipindahkan ke folder dataset_model:\n","Training: /content/dataset_model/training\n","Validation: /content/dataset_model/validation\n","Isi folder training: ['anorganik', 'organik']\n","Isi folder validation: ['anorganik', 'organik']\n","Found 800 images belonging to 2 classes.\n","Found 200 images belonging to 2 classes.\n"]}]},{"cell_type":"code","source":["# Image Augmentation\n","data_augmentation = tf.keras.Sequential([\n","    tf.keras.layers.Input(shape=(IMAGE_SIZE[0], IMAGE_SIZE[1], 3)),  # Menggunakan Input layer\n","    tf.keras.layers.RandomFlip(\"horizontal\"),\n","    tf.keras.layers.RandomRotation(0.1),\n","    tf.keras.layers.RandomZoom(0.1),\n","    tf.keras.layers.Rescaling(1./255)\n","])\n"],"metadata":{"id":"OYh1Ur0EeKRb"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["# **Modeling**"],"metadata":{"id":"b1aRy2QIpBKn"}},{"cell_type":"markdown","source":["# **Membuat Arsitektur CNN**\n","\n"],"metadata":{"id":"qn1XgdoZi9A4"}},{"cell_type":"markdown","source":["**Penyusunan Layer**"],"metadata":{"id":"PqBAL4BijLp9"}},{"cell_type":"code","source":["cnn_model = tf.keras.models.Sequential([\n","    data_augmentation,\n","    tf.keras.layers.Conv2D(32, 3, padding='same', activation='relu'),\n","    tf.keras.layers.MaxPooling2D(),\n","    tf.keras.layers.Conv2D(64, 3, padding='same', activation='relu'),\n","    tf.keras.layers.MaxPooling2D(),\n","    tf.keras.layers.Conv2D(128, 3, padding='same', activation='relu'),\n","    tf.keras.layers.MaxPooling2D(),\n","    tf.keras.layers.Dropout(0.3),\n","    tf.keras.layers.Flatten(),\n","    tf.keras.layers.Dense(128, activation='relu'),\n","    tf.keras.layers.Dense(64, activation='relu'),\n","    tf.keras.layers.Dense(1, activation='sigmoid')  # Satu neuron dengan sigmoid untuk binary_crossentropy\n","])\n","\n","cnn_model.compile(\n","    loss='binary_crossentropy',  # binary_crossentropy untuk output biner\n","    optimizer=tf.keras.optimizers.Adam(learning_rate=0.0001),\n","    metrics=['accuracy']\n",")\n"],"metadata":{"id":"Fw2CNfQQpEz8"},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":["**Melatih Model CNN**"],"metadata":{"id":"Y4fU6meUrQw1"}},{"cell_type":"code","source":["# Training model CNN\n","cnn_hist = cnn_model.fit(\n","    train_data,\n","    epochs=256,\n","    validation_data = valid_data\n",")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"GBMe69kUrQDN","executionInfo":{"status":"ok","timestamp":1736930331580,"user_tz":-420,"elapsed":1037024,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"1c4db1c2-497e-416b-f4ff-a48e5e93b953"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Epoch 1/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 113ms/step - accuracy: 0.4999 - loss: 0.6929 - val_accuracy: 0.5000 - val_loss: 0.6899\n","Epoch 2/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 100ms/step - accuracy: 0.5564 - loss: 0.6876 - val_accuracy: 0.7850 - val_loss: 0.6750\n","Epoch 3/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.6864 - loss: 0.6653 - val_accuracy: 0.6950 - val_loss: 0.6225\n","Epoch 4/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 95ms/step - accuracy: 0.7140 - loss: 0.6275 - val_accuracy: 0.7600 - val_loss: 0.5835\n","Epoch 5/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 95ms/step - accuracy: 0.7759 - loss: 0.5802 - val_accuracy: 0.7500 - val_loss: 0.5438\n","Epoch 6/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.7323 - loss: 0.5661 - val_accuracy: 0.7600 - val_loss: 0.5205\n","Epoch 7/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.7590 - loss: 0.5429 - val_accuracy: 0.8250 - val_loss: 0.4830\n","Epoch 8/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 104ms/step - accuracy: 0.8143 - loss: 0.4885 - val_accuracy: 0.7950 - val_loss: 0.4876\n","Epoch 9/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.7907 - loss: 0.4922 - val_accuracy: 0.7500 - val_loss: 0.5048\n","Epoch 10/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 102ms/step - accuracy: 0.7948 - loss: 0.4790 - val_accuracy: 0.7200 - val_loss: 0.5316\n","Epoch 11/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.7972 - loss: 0.4975 - val_accuracy: 0.7850 - val_loss: 0.4792\n","Epoch 12/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 96ms/step - accuracy: 0.7657 - loss: 0.4785 - val_accuracy: 0.8000 - val_loss: 0.4475\n","Epoch 13/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8159 - loss: 0.4465 - val_accuracy: 0.8300 - val_loss: 0.4336\n","Epoch 14/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 95ms/step - accuracy: 0.8092 - loss: 0.4525 - val_accuracy: 0.8300 - val_loss: 0.4281\n","Epoch 15/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 102ms/step - accuracy: 0.8046 - loss: 0.4816 - val_accuracy: 0.8250 - val_loss: 0.4216\n","Epoch 16/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 105ms/step - accuracy: 0.8133 - loss: 0.4318 - val_accuracy: 0.7550 - val_loss: 0.5160\n","Epoch 17/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.7785 - loss: 0.5022 - val_accuracy: 0.8100 - val_loss: 0.4240\n","Epoch 18/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 110ms/step - accuracy: 0.8389 - loss: 0.3965 - val_accuracy: 0.7850 - val_loss: 0.4376\n","Epoch 19/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 103ms/step - accuracy: 0.7947 - loss: 0.4622 - val_accuracy: 0.8200 - val_loss: 0.4154\n","Epoch 20/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8093 - loss: 0.4389 - val_accuracy: 0.8150 - val_loss: 0.4254\n","Epoch 21/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8216 - loss: 0.4323 - val_accuracy: 0.8400 - val_loss: 0.3993\n","Epoch 22/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 113ms/step - accuracy: 0.7961 - loss: 0.4557 - val_accuracy: 0.8150 - val_loss: 0.4080\n","Epoch 23/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 96ms/step - accuracy: 0.7825 - loss: 0.4578 - val_accuracy: 0.8150 - val_loss: 0.4025\n","Epoch 24/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.8191 - loss: 0.4138 - val_accuracy: 0.8300 - val_loss: 0.4068\n","Epoch 25/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.8125 - loss: 0.4321 - val_accuracy: 0.7600 - val_loss: 0.5068\n","Epoch 26/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8217 - loss: 0.4203 - val_accuracy: 0.8250 - val_loss: 0.3947\n","Epoch 27/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8217 - loss: 0.4098 - val_accuracy: 0.8100 - val_loss: 0.4115\n","Epoch 28/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 112ms/step - accuracy: 0.8247 - loss: 0.4277 - val_accuracy: 0.8200 - val_loss: 0.3878\n","Epoch 29/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 102ms/step - accuracy: 0.8499 - loss: 0.3501 - val_accuracy: 0.8100 - val_loss: 0.4072\n","Epoch 30/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8376 - loss: 0.3926 - val_accuracy: 0.8250 - val_loss: 0.4132\n","Epoch 31/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8384 - loss: 0.3968 - val_accuracy: 0.8200 - val_loss: 0.3876\n","Epoch 32/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.8596 - loss: 0.3721 - val_accuracy: 0.8350 - val_loss: 0.3676\n","Epoch 33/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8422 - loss: 0.3841 - val_accuracy: 0.8400 - val_loss: 0.3492\n","Epoch 34/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.8893 - loss: 0.3457 - val_accuracy: 0.7900 - val_loss: 0.4408\n","Epoch 35/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 101ms/step - accuracy: 0.8000 - loss: 0.4530 - val_accuracy: 0.7550 - val_loss: 0.5647\n","Epoch 36/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 100ms/step - accuracy: 0.8290 - loss: 0.3796 - val_accuracy: 0.8450 - val_loss: 0.3633\n","Epoch 37/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.8578 - loss: 0.3788 - val_accuracy: 0.8250 - val_loss: 0.3716\n","Epoch 38/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 96ms/step - accuracy: 0.8578 - loss: 0.3438 - val_accuracy: 0.8450 - val_loss: 0.3725\n","Epoch 39/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8563 - loss: 0.3571 - val_accuracy: 0.8350 - val_loss: 0.3613\n","Epoch 40/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8580 - loss: 0.3634 - val_accuracy: 0.8600 - val_loss: 0.3372\n","Epoch 41/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 111ms/step - accuracy: 0.8561 - loss: 0.3764 - val_accuracy: 0.8350 - val_loss: 0.3760\n","Epoch 42/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 111ms/step - accuracy: 0.8811 - loss: 0.3081 - val_accuracy: 0.8250 - val_loss: 0.4162\n","Epoch 43/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.8696 - loss: 0.3372 - val_accuracy: 0.8750 - val_loss: 0.3250\n","Epoch 44/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 115ms/step - accuracy: 0.8471 - loss: 0.3929 - val_accuracy: 0.8450 - val_loss: 0.3462\n","Epoch 45/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8758 - loss: 0.3550 - val_accuracy: 0.8550 - val_loss: 0.3256\n","Epoch 46/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.8356 - loss: 0.3948 - val_accuracy: 0.8400 - val_loss: 0.3335\n","Epoch 47/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 121ms/step - accuracy: 0.8688 - loss: 0.3491 - val_accuracy: 0.8350 - val_loss: 0.3436\n","Epoch 48/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8371 - loss: 0.3619 - val_accuracy: 0.8300 - val_loss: 0.3747\n","Epoch 49/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8628 - loss: 0.3574 - val_accuracy: 0.8750 - val_loss: 0.3091\n","Epoch 50/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.8739 - loss: 0.3471 - val_accuracy: 0.8250 - val_loss: 0.4306\n","Epoch 51/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 111ms/step - accuracy: 0.8543 - loss: 0.3889 - val_accuracy: 0.8500 - val_loss: 0.3140\n","Epoch 52/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.8265 - loss: 0.3985 - val_accuracy: 0.8300 - val_loss: 0.3620\n","Epoch 53/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8250 - loss: 0.4179 - val_accuracy: 0.8650 - val_loss: 0.3282\n","Epoch 54/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 117ms/step - accuracy: 0.8533 - loss: 0.3901 - val_accuracy: 0.8550 - val_loss: 0.3140\n","Epoch 55/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.8736 - loss: 0.3283 - val_accuracy: 0.8800 - val_loss: 0.3008\n","Epoch 56/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8853 - loss: 0.3044 - val_accuracy: 0.8850 - val_loss: 0.2946\n","Epoch 57/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 101ms/step - accuracy: 0.8459 - loss: 0.3737 - val_accuracy: 0.8600 - val_loss: 0.3021\n","Epoch 58/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 104ms/step - accuracy: 0.8776 - loss: 0.3427 - val_accuracy: 0.8300 - val_loss: 0.4505\n","Epoch 59/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.8577 - loss: 0.3454 - val_accuracy: 0.8800 - val_loss: 0.2996\n","Epoch 60/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.8587 - loss: 0.3452 - val_accuracy: 0.8050 - val_loss: 0.4429\n","Epoch 61/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 102ms/step - accuracy: 0.8428 - loss: 0.3641 - val_accuracy: 0.8300 - val_loss: 0.3767\n","Epoch 62/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 102ms/step - accuracy: 0.8382 - loss: 0.4017 - val_accuracy: 0.8600 - val_loss: 0.3395\n","Epoch 63/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8817 - loss: 0.2835 - val_accuracy: 0.8500 - val_loss: 0.3210\n","Epoch 64/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8852 - loss: 0.3052 - val_accuracy: 0.8950 - val_loss: 0.2962\n","Epoch 65/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 113ms/step - accuracy: 0.8830 - loss: 0.2928 - val_accuracy: 0.8850 - val_loss: 0.2861\n","Epoch 66/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.8834 - loss: 0.3023 - val_accuracy: 0.8250 - val_loss: 0.3906\n","Epoch 67/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8719 - loss: 0.3192 - val_accuracy: 0.8450 - val_loss: 0.3229\n","Epoch 68/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 117ms/step - accuracy: 0.8582 - loss: 0.3294 - val_accuracy: 0.8700 - val_loss: 0.2988\n","Epoch 69/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.8725 - loss: 0.3185 - val_accuracy: 0.8650 - val_loss: 0.3232\n","Epoch 70/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.8910 - loss: 0.2874 - val_accuracy: 0.8850 - val_loss: 0.2984\n","Epoch 71/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.8760 - loss: 0.3013 - val_accuracy: 0.8600 - val_loss: 0.3123\n","Epoch 72/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.8879 - loss: 0.3028 - val_accuracy: 0.8600 - val_loss: 0.3248\n","Epoch 73/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 103ms/step - accuracy: 0.8647 - loss: 0.3271 - val_accuracy: 0.8700 - val_loss: 0.2926\n","Epoch 74/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.8952 - loss: 0.2938 - val_accuracy: 0.8350 - val_loss: 0.3597\n","Epoch 75/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8603 - loss: 0.3610 - val_accuracy: 0.8800 - val_loss: 0.2789\n","Epoch 76/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 113ms/step - accuracy: 0.8792 - loss: 0.2887 - val_accuracy: 0.8700 - val_loss: 0.2713\n","Epoch 77/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 100ms/step - accuracy: 0.8855 - loss: 0.2928 - val_accuracy: 0.8950 - val_loss: 0.2580\n","Epoch 78/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 112ms/step - accuracy: 0.8994 - loss: 0.2623 - val_accuracy: 0.8750 - val_loss: 0.3088\n","Epoch 79/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 108ms/step - accuracy: 0.8468 - loss: 0.3324 - val_accuracy: 0.8300 - val_loss: 0.3735\n","Epoch 80/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.8694 - loss: 0.2931 - val_accuracy: 0.9000 - val_loss: 0.2497\n","Epoch 81/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.8940 - loss: 0.2798 - val_accuracy: 0.9000 - val_loss: 0.2571\n","Epoch 82/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9110 - loss: 0.2527 - val_accuracy: 0.8050 - val_loss: 0.4270\n","Epoch 83/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 111ms/step - accuracy: 0.8204 - loss: 0.4002 - val_accuracy: 0.8750 - val_loss: 0.2948\n","Epoch 84/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.8727 - loss: 0.3278 - val_accuracy: 0.8800 - val_loss: 0.2965\n","Epoch 85/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.8891 - loss: 0.3094 - val_accuracy: 0.8900 - val_loss: 0.2784\n","Epoch 86/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8770 - loss: 0.3044 - val_accuracy: 0.8950 - val_loss: 0.2607\n","Epoch 87/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 116ms/step - accuracy: 0.8859 - loss: 0.2980 - val_accuracy: 0.8800 - val_loss: 0.2682\n","Epoch 88/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9023 - loss: 0.2772 - val_accuracy: 0.8750 - val_loss: 0.2919\n","Epoch 89/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8897 - loss: 0.2808 - val_accuracy: 0.8650 - val_loss: 0.2955\n","Epoch 90/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 100ms/step - accuracy: 0.8673 - loss: 0.3039 - val_accuracy: 0.8400 - val_loss: 0.3850\n","Epoch 91/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8705 - loss: 0.3136 - val_accuracy: 0.8800 - val_loss: 0.2814\n","Epoch 92/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 95ms/step - accuracy: 0.8838 - loss: 0.3076 - val_accuracy: 0.8800 - val_loss: 0.2680\n","Epoch 93/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 114ms/step - accuracy: 0.8872 - loss: 0.2978 - val_accuracy: 0.8600 - val_loss: 0.3357\n","Epoch 94/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 99ms/step - accuracy: 0.9095 - loss: 0.2503 - val_accuracy: 0.8900 - val_loss: 0.2675\n","Epoch 95/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9172 - loss: 0.2464 - val_accuracy: 0.9250 - val_loss: 0.2320\n","Epoch 96/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 104ms/step - accuracy: 0.8875 - loss: 0.2708 - val_accuracy: 0.9000 - val_loss: 0.2282\n","Epoch 97/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.8991 - loss: 0.2673 - val_accuracy: 0.9100 - val_loss: 0.2251\n","Epoch 98/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8958 - loss: 0.2643 - val_accuracy: 0.9200 - val_loss: 0.2456\n","Epoch 99/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8894 - loss: 0.2561 - val_accuracy: 0.8500 - val_loss: 0.3258\n","Epoch 100/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 104ms/step - accuracy: 0.9101 - loss: 0.2181 - val_accuracy: 0.9000 - val_loss: 0.2638\n","Epoch 101/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8879 - loss: 0.3234 - val_accuracy: 0.9300 - val_loss: 0.2313\n","Epoch 102/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9105 - loss: 0.2421 - val_accuracy: 0.8750 - val_loss: 0.2888\n","Epoch 103/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 101ms/step - accuracy: 0.9231 - loss: 0.2417 - val_accuracy: 0.9150 - val_loss: 0.2191\n","Epoch 104/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 96ms/step - accuracy: 0.9046 - loss: 0.2348 - val_accuracy: 0.9200 - val_loss: 0.2160\n","Epoch 105/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 95ms/step - accuracy: 0.8951 - loss: 0.2675 - val_accuracy: 0.8750 - val_loss: 0.2862\n","Epoch 106/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 115ms/step - accuracy: 0.9209 - loss: 0.2492 - val_accuracy: 0.9200 - val_loss: 0.2312\n","Epoch 107/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.8778 - loss: 0.2968 - val_accuracy: 0.9300 - val_loss: 0.2110\n","Epoch 108/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9131 - loss: 0.2435 - val_accuracy: 0.9300 - val_loss: 0.2076\n","Epoch 109/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.8948 - loss: 0.2694 - val_accuracy: 0.8650 - val_loss: 0.3289\n","Epoch 110/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 120ms/step - accuracy: 0.8943 - loss: 0.2634 - val_accuracy: 0.9050 - val_loss: 0.2627\n","Epoch 111/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 98ms/step - accuracy: 0.9119 - loss: 0.2464 - val_accuracy: 0.9150 - val_loss: 0.2236\n","Epoch 112/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 110ms/step - accuracy: 0.9081 - loss: 0.2435 - val_accuracy: 0.9250 - val_loss: 0.2018\n","Epoch 113/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 119ms/step - accuracy: 0.9021 - loss: 0.2537 - val_accuracy: 0.9150 - val_loss: 0.2061\n","Epoch 114/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9121 - loss: 0.2512 - val_accuracy: 0.9150 - val_loss: 0.2434\n","Epoch 115/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.8978 - loss: 0.2429 - val_accuracy: 0.9400 - val_loss: 0.2004\n","Epoch 116/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 117ms/step - accuracy: 0.9066 - loss: 0.2440 - val_accuracy: 0.9350 - val_loss: 0.1981\n","Epoch 117/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9142 - loss: 0.2439 - val_accuracy: 0.9350 - val_loss: 0.2060\n","Epoch 118/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9005 - loss: 0.2425 - val_accuracy: 0.9300 - val_loss: 0.2129\n","Epoch 119/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 112ms/step - accuracy: 0.8911 - loss: 0.2451 - val_accuracy: 0.9100 - val_loss: 0.2556\n","Epoch 120/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 100ms/step - accuracy: 0.8982 - loss: 0.2634 - val_accuracy: 0.9200 - val_loss: 0.2381\n","Epoch 121/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9014 - loss: 0.2684 - val_accuracy: 0.9400 - val_loss: 0.2028\n","Epoch 122/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 108ms/step - accuracy: 0.9282 - loss: 0.2096 - val_accuracy: 0.9300 - val_loss: 0.2110\n","Epoch 123/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 104ms/step - accuracy: 0.9331 - loss: 0.1955 - val_accuracy: 0.9000 - val_loss: 0.2581\n","Epoch 124/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9133 - loss: 0.2153 - val_accuracy: 0.8800 - val_loss: 0.3213\n","Epoch 125/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9151 - loss: 0.2373 - val_accuracy: 0.9050 - val_loss: 0.2464\n","Epoch 126/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9076 - loss: 0.2320 - val_accuracy: 0.9150 - val_loss: 0.2580\n","Epoch 127/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9138 - loss: 0.1997 - val_accuracy: 0.9400 - val_loss: 0.2012\n","Epoch 128/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 111ms/step - accuracy: 0.8981 - loss: 0.2193 - val_accuracy: 0.9000 - val_loss: 0.2773\n","Epoch 129/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 103ms/step - accuracy: 0.8684 - loss: 0.3171 - val_accuracy: 0.9250 - val_loss: 0.2141\n","Epoch 130/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.8975 - loss: 0.2763 - val_accuracy: 0.9000 - val_loss: 0.2547\n","Epoch 131/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9362 - loss: 0.1918 - val_accuracy: 0.8800 - val_loss: 0.3197\n","Epoch 132/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9240 - loss: 0.2306 - val_accuracy: 0.9350 - val_loss: 0.1905\n","Epoch 133/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9307 - loss: 0.2153 - val_accuracy: 0.8150 - val_loss: 0.4301\n","Epoch 134/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9197 - loss: 0.2124 - val_accuracy: 0.9250 - val_loss: 0.2040\n","Epoch 135/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9429 - loss: 0.1847 - val_accuracy: 0.9150 - val_loss: 0.2386\n","Epoch 136/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 102ms/step - accuracy: 0.9052 - loss: 0.2231 - val_accuracy: 0.9250 - val_loss: 0.2072\n","Epoch 137/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9120 - loss: 0.2269 - val_accuracy: 0.8900 - val_loss: 0.2998\n","Epoch 138/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 103ms/step - accuracy: 0.9092 - loss: 0.2229 - val_accuracy: 0.9400 - val_loss: 0.1881\n","Epoch 139/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9244 - loss: 0.1946 - val_accuracy: 0.9250 - val_loss: 0.2115\n","Epoch 140/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9458 - loss: 0.1907 - val_accuracy: 0.9250 - val_loss: 0.2108\n","Epoch 141/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9157 - loss: 0.1998 - val_accuracy: 0.9300 - val_loss: 0.1928\n","Epoch 142/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 107ms/step - accuracy: 0.9117 - loss: 0.2265 - val_accuracy: 0.9200 - val_loss: 0.1855\n","Epoch 143/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.8924 - loss: 0.2638 - val_accuracy: 0.9350 - val_loss: 0.1951\n","Epoch 144/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9279 - loss: 0.2084 - val_accuracy: 0.9100 - val_loss: 0.2466\n","Epoch 145/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9166 - loss: 0.2114 - val_accuracy: 0.8150 - val_loss: 0.4711\n","Epoch 146/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.8831 - loss: 0.2860 - val_accuracy: 0.8400 - val_loss: 0.4274\n","Epoch 147/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 102ms/step - accuracy: 0.8387 - loss: 0.3500 - val_accuracy: 0.9350 - val_loss: 0.1925\n","Epoch 148/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9091 - loss: 0.2299 - val_accuracy: 0.9300 - val_loss: 0.2071\n","Epoch 149/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9161 - loss: 0.2191 - val_accuracy: 0.9200 - val_loss: 0.2314\n","Epoch 150/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 105ms/step - accuracy: 0.9309 - loss: 0.2041 - val_accuracy: 0.9050 - val_loss: 0.2594\n","Epoch 151/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 99ms/step - accuracy: 0.9359 - loss: 0.1939 - val_accuracy: 0.9000 - val_loss: 0.2852\n","Epoch 152/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9157 - loss: 0.1958 - val_accuracy: 0.9350 - val_loss: 0.1884\n","Epoch 153/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9180 - loss: 0.2052 - val_accuracy: 0.8700 - val_loss: 0.2467\n","Epoch 154/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 105ms/step - accuracy: 0.8948 - loss: 0.2955 - val_accuracy: 0.8550 - val_loss: 0.3783\n","Epoch 155/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.8818 - loss: 0.2830 - val_accuracy: 0.9250 - val_loss: 0.2232\n","Epoch 156/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9256 - loss: 0.2071 - val_accuracy: 0.9400 - val_loss: 0.1935\n","Epoch 157/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 120ms/step - accuracy: 0.9337 - loss: 0.1955 - val_accuracy: 0.9000 - val_loss: 0.2608\n","Epoch 158/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9182 - loss: 0.2015 - val_accuracy: 0.8900 - val_loss: 0.3075\n","Epoch 159/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9491 - loss: 0.1911 - val_accuracy: 0.9450 - val_loss: 0.1786\n","Epoch 160/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 101ms/step - accuracy: 0.9247 - loss: 0.2061 - val_accuracy: 0.9300 - val_loss: 0.1898\n","Epoch 161/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 112ms/step - accuracy: 0.9098 - loss: 0.2434 - val_accuracy: 0.9250 - val_loss: 0.1973\n","Epoch 162/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 112ms/step - accuracy: 0.9192 - loss: 0.1954 - val_accuracy: 0.9350 - val_loss: 0.1848\n","Epoch 163/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 107ms/step - accuracy: 0.9209 - loss: 0.2360 - val_accuracy: 0.8950 - val_loss: 0.2919\n","Epoch 164/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9266 - loss: 0.1839 - val_accuracy: 0.9200 - val_loss: 0.2169\n","Epoch 165/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 111ms/step - accuracy: 0.9235 - loss: 0.1993 - val_accuracy: 0.8900 - val_loss: 0.2741\n","Epoch 166/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9356 - loss: 0.1877 - val_accuracy: 0.9300 - val_loss: 0.1963\n","Epoch 167/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 106ms/step - accuracy: 0.9243 - loss: 0.2142 - val_accuracy: 0.9200 - val_loss: 0.2171\n","Epoch 168/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 111ms/step - accuracy: 0.9216 - loss: 0.1939 - val_accuracy: 0.9450 - val_loss: 0.1708\n","Epoch 169/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9260 - loss: 0.1992 - val_accuracy: 0.9200 - val_loss: 0.2241\n","Epoch 170/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 120ms/step - accuracy: 0.9335 - loss: 0.1839 - val_accuracy: 0.8950 - val_loss: 0.2725\n","Epoch 171/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 98ms/step - accuracy: 0.9234 - loss: 0.1774 - val_accuracy: 0.9400 - val_loss: 0.1793\n","Epoch 172/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9330 - loss: 0.1825 - val_accuracy: 0.9250 - val_loss: 0.2064\n","Epoch 173/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 119ms/step - accuracy: 0.9093 - loss: 0.2076 - val_accuracy: 0.9300 - val_loss: 0.1779\n","Epoch 174/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9376 - loss: 0.1672 - val_accuracy: 0.9300 - val_loss: 0.1876\n","Epoch 175/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9147 - loss: 0.2420 - val_accuracy: 0.8550 - val_loss: 0.4137\n","Epoch 176/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9096 - loss: 0.2019 - val_accuracy: 0.9150 - val_loss: 0.2297\n","Epoch 177/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 103ms/step - accuracy: 0.9404 - loss: 0.1619 - val_accuracy: 0.9250 - val_loss: 0.2170\n","Epoch 178/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9318 - loss: 0.1737 - val_accuracy: 0.9400 - val_loss: 0.1841\n","Epoch 179/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 110ms/step - accuracy: 0.9226 - loss: 0.2057 - val_accuracy: 0.9350 - val_loss: 0.1946\n","Epoch 180/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9321 - loss: 0.1881 - val_accuracy: 0.9200 - val_loss: 0.2146\n","Epoch 181/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9283 - loss: 0.1827 - val_accuracy: 0.9200 - val_loss: 0.2567\n","Epoch 182/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 111ms/step - accuracy: 0.9266 - loss: 0.1681 - val_accuracy: 0.9150 - val_loss: 0.2655\n","Epoch 183/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9329 - loss: 0.2062 - val_accuracy: 0.8600 - val_loss: 0.3959\n","Epoch 184/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9211 - loss: 0.2330 - val_accuracy: 0.9400 - val_loss: 0.1828\n","Epoch 185/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9345 - loss: 0.1895 - val_accuracy: 0.9150 - val_loss: 0.2414\n","Epoch 186/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9409 - loss: 0.1744 - val_accuracy: 0.9500 - val_loss: 0.1636\n","Epoch 187/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 106ms/step - accuracy: 0.9151 - loss: 0.1999 - val_accuracy: 0.9250 - val_loss: 0.2015\n","Epoch 188/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9290 - loss: 0.1655 - val_accuracy: 0.9350 - val_loss: 0.1968\n","Epoch 189/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9387 - loss: 0.1682 - val_accuracy: 0.9150 - val_loss: 0.2405\n","Epoch 190/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 106ms/step - accuracy: 0.9372 - loss: 0.1763 - val_accuracy: 0.9100 - val_loss: 0.2632\n","Epoch 191/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 99ms/step - accuracy: 0.9248 - loss: 0.1879 - val_accuracy: 0.9450 - val_loss: 0.1787\n","Epoch 192/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9331 - loss: 0.1982 - val_accuracy: 0.9250 - val_loss: 0.2032\n","Epoch 193/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9432 - loss: 0.1857 - val_accuracy: 0.8900 - val_loss: 0.2935\n","Epoch 194/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9227 - loss: 0.1812 - val_accuracy: 0.8700 - val_loss: 0.3565\n","Epoch 195/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9411 - loss: 0.1640 - val_accuracy: 0.9200 - val_loss: 0.2158\n","Epoch 196/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 103ms/step - accuracy: 0.9252 - loss: 0.1915 - val_accuracy: 0.9250 - val_loss: 0.2004\n","Epoch 197/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9273 - loss: 0.1657 - val_accuracy: 0.8950 - val_loss: 0.2789\n","Epoch 198/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 105ms/step - accuracy: 0.9246 - loss: 0.1994 - val_accuracy: 0.9400 - val_loss: 0.1644\n","Epoch 199/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 106ms/step - accuracy: 0.9126 - loss: 0.2061 - val_accuracy: 0.8950 - val_loss: 0.2813\n","Epoch 200/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9461 - loss: 0.1525 - val_accuracy: 0.9450 - val_loss: 0.1849\n","Epoch 201/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9512 - loss: 0.1636 - val_accuracy: 0.9450 - val_loss: 0.1685\n","Epoch 202/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9280 - loss: 0.1889 - val_accuracy: 0.8950 - val_loss: 0.3035\n","Epoch 203/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9395 - loss: 0.1657 - val_accuracy: 0.9350 - val_loss: 0.1760\n","Epoch 204/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9288 - loss: 0.1883 - val_accuracy: 0.9400 - val_loss: 0.1604\n","Epoch 205/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 110ms/step - accuracy: 0.9372 - loss: 0.1771 - val_accuracy: 0.9450 - val_loss: 0.1782\n","Epoch 206/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 113ms/step - accuracy: 0.9337 - loss: 0.1684 - val_accuracy: 0.9400 - val_loss: 0.1772\n","Epoch 207/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 112ms/step - accuracy: 0.9349 - loss: 0.1504 - val_accuracy: 0.9300 - val_loss: 0.1979\n","Epoch 208/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 106ms/step - accuracy: 0.9369 - loss: 0.1768 - val_accuracy: 0.9350 - val_loss: 0.1683\n","Epoch 209/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9415 - loss: 0.1642 - val_accuracy: 0.9350 - val_loss: 0.1893\n","Epoch 210/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 111ms/step - accuracy: 0.9431 - loss: 0.1726 - val_accuracy: 0.9200 - val_loss: 0.2106\n","Epoch 211/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9324 - loss: 0.1759 - val_accuracy: 0.9350 - val_loss: 0.1700\n","Epoch 212/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9209 - loss: 0.1871 - val_accuracy: 0.8600 - val_loss: 0.4004\n","Epoch 213/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 116ms/step - accuracy: 0.9128 - loss: 0.2225 - val_accuracy: 0.9450 - val_loss: 0.1795\n","Epoch 214/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9412 - loss: 0.1649 - val_accuracy: 0.9300 - val_loss: 0.1758\n","Epoch 215/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9386 - loss: 0.1682 - val_accuracy: 0.9150 - val_loss: 0.2447\n","Epoch 216/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 108ms/step - accuracy: 0.9419 - loss: 0.1495 - val_accuracy: 0.9050 - val_loss: 0.2587\n","Epoch 217/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9336 - loss: 0.1708 - val_accuracy: 0.9350 - val_loss: 0.2074\n","Epoch 218/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9372 - loss: 0.1818 - val_accuracy: 0.9200 - val_loss: 0.2299\n","Epoch 219/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9495 - loss: 0.1457 - val_accuracy: 0.9400 - val_loss: 0.1880\n","Epoch 220/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9451 - loss: 0.1470 - val_accuracy: 0.9450 - val_loss: 0.1836\n","Epoch 221/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9334 - loss: 0.1650 - val_accuracy: 0.9350 - val_loss: 0.1532\n","Epoch 222/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 104ms/step - accuracy: 0.9319 - loss: 0.1894 - val_accuracy: 0.9450 - val_loss: 0.1731\n","Epoch 223/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9236 - loss: 0.1748 - val_accuracy: 0.9300 - val_loss: 0.2062\n","Epoch 224/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9582 - loss: 0.1449 - val_accuracy: 0.9450 - val_loss: 0.1813\n","Epoch 225/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 108ms/step - accuracy: 0.9364 - loss: 0.1750 - val_accuracy: 0.9400 - val_loss: 0.1871\n","Epoch 226/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9294 - loss: 0.1678 - val_accuracy: 0.8650 - val_loss: 0.3514\n","Epoch 227/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9490 - loss: 0.1414 - val_accuracy: 0.9500 - val_loss: 0.1845\n","Epoch 228/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 107ms/step - accuracy: 0.9346 - loss: 0.1710 - val_accuracy: 0.9450 - val_loss: 0.1918\n","Epoch 229/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9431 - loss: 0.1593 - val_accuracy: 0.9450 - val_loss: 0.1659\n","Epoch 230/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9334 - loss: 0.1868 - val_accuracy: 0.9000 - val_loss: 0.2738\n","Epoch 231/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 95ms/step - accuracy: 0.9389 - loss: 0.1626 - val_accuracy: 0.9350 - val_loss: 0.2096\n","Epoch 232/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9278 - loss: 0.1709 - val_accuracy: 0.9450 - val_loss: 0.1778\n","Epoch 233/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9455 - loss: 0.1597 - val_accuracy: 0.9000 - val_loss: 0.2703\n","Epoch 234/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 100ms/step - accuracy: 0.9371 - loss: 0.1672 - val_accuracy: 0.9200 - val_loss: 0.1947\n","Epoch 235/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 96ms/step - accuracy: 0.9454 - loss: 0.1605 - val_accuracy: 0.9500 - val_loss: 0.1695\n","Epoch 236/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 105ms/step - accuracy: 0.9347 - loss: 0.1670 - val_accuracy: 0.9350 - val_loss: 0.1937\n","Epoch 237/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 99ms/step - accuracy: 0.9337 - loss: 0.1579 - val_accuracy: 0.9000 - val_loss: 0.2682\n","Epoch 238/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9569 - loss: 0.1398 - val_accuracy: 0.9350 - val_loss: 0.1894\n","Epoch 239/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 98ms/step - accuracy: 0.9441 - loss: 0.1319 - val_accuracy: 0.9350 - val_loss: 0.1925\n","Epoch 240/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 107ms/step - accuracy: 0.9262 - loss: 0.1812 - val_accuracy: 0.9000 - val_loss: 0.2831\n","Epoch 241/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 100ms/step - accuracy: 0.9498 - loss: 0.1280 - val_accuracy: 0.8800 - val_loss: 0.3107\n","Epoch 242/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 97ms/step - accuracy: 0.9410 - loss: 0.1758 - val_accuracy: 0.9050 - val_loss: 0.2682\n","Epoch 243/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 112ms/step - accuracy: 0.9313 - loss: 0.1643 - val_accuracy: 0.9450 - val_loss: 0.1696\n","Epoch 244/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9436 - loss: 0.1498 - val_accuracy: 0.9450 - val_loss: 0.1753\n","Epoch 245/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9652 - loss: 0.1188 - val_accuracy: 0.9450 - val_loss: 0.1694\n","Epoch 246/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9485 - loss: 0.1554 - val_accuracy: 0.9300 - val_loss: 0.1997\n","Epoch 247/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9385 - loss: 0.1550 - val_accuracy: 0.9400 - val_loss: 0.1995\n","Epoch 248/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 96ms/step - accuracy: 0.9477 - loss: 0.1302 - val_accuracy: 0.9050 - val_loss: 0.2558\n","Epoch 249/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m6s\u001b[0m 111ms/step - accuracy: 0.9565 - loss: 0.1244 - val_accuracy: 0.9250 - val_loss: 0.2225\n","Epoch 250/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9503 - loss: 0.1297 - val_accuracy: 0.9450 - val_loss: 0.1694\n","Epoch 251/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 98ms/step - accuracy: 0.9614 - loss: 0.1174 - val_accuracy: 0.9500 - val_loss: 0.1856\n","Epoch 252/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9544 - loss: 0.1271 - val_accuracy: 0.9400 - val_loss: 0.1606\n","Epoch 253/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 99ms/step - accuracy: 0.9448 - loss: 0.1399 - val_accuracy: 0.9300 - val_loss: 0.1685\n","Epoch 254/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 105ms/step - accuracy: 0.9339 - loss: 0.1791 - val_accuracy: 0.8950 - val_loss: 0.2981\n","Epoch 255/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 111ms/step - accuracy: 0.9445 - loss: 0.1278 - val_accuracy: 0.9100 - val_loss: 0.2289\n","Epoch 256/256\n","\u001b[1m25/25\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 97ms/step - accuracy: 0.9489 - loss: 0.1485 - val_accuracy: 0.9400 - val_loss: 0.1729\n"]}]},{"cell_type":"code","source":["# Membuat plot akurasi model CNN\n","plt.figure(figsize=(10,4))\n","plt.plot(cnn_hist.history['accuracy'])\n","plt.plot(cnn_hist.history['val_accuracy'])\n","plt.title('CNN model accuracy')\n","plt.ylabel('accuracy')\n","plt.xlabel('epoch')\n","plt.legend(['train', 'test'], loc='upper left')\n","plt.grid(True)\n","plt.show()\n","\n","print()\n","\n","# Membuat plot loss model CNN\n","plt.figure(figsize=(10,4))\n","plt.plot(cnn_hist.history['loss'])\n","plt.plot(cnn_hist.history['val_loss'])\n","plt.title('CNN model loss')\n","plt.ylabel('loss')\n","plt.xlabel('epoch')\n","plt.legend(['train', 'test'], loc='upper left')\n","plt.grid(True)\n","plt.show()\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/","height":820},"id":"V3QvL6k2vUoc","executionInfo":{"status":"ok","timestamp":1736931021425,"user_tz":-420,"elapsed":1656,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"93acfa07-7dc4-40bc-f402-b40acca2e0cf"},"execution_count":null,"outputs":[{"output_type":"display_data","data":{"text/plain":["<Figure size 1000x400 with 1 Axes>"],"image/png":"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\n"},"metadata":{}},{"output_type":"stream","name":"stdout","text":["\n"]},{"output_type":"display_data","data":{"text/plain":["<Figure size 1000x400 with 1 Axes>"],"image/png":"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\n"},"metadata":{}}]},{"cell_type":"code","source":["from sklearn.metrics import f1_score\n","\n","# Calculate overall training and evaluation percentages\n","train_acc = cnn_hist.history['accuracy'][-1]\n","val_acc = cnn_hist.history['val_accuracy'][-1]\n","\n","print(f\"\\nOverall Training Accuracy: {train_acc * 100:.2f}%\")\n","print(f\"Overall Validation Accuracy: {val_acc * 100:.2f}%\")"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"g0kmJQRkKETY","executionInfo":{"status":"ok","timestamp":1736931144619,"user_tz":-420,"elapsed":330,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"16a5b8ee-3e97-4bbd-9258-a70ce677de6b"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["\n","Overall Training Accuracy: 94.88%\n","Overall Validation Accuracy: 94.00%\n"]}]},{"cell_type":"code","source":["# Import library TensorFlow\n","import tensorflow as tf\n","\n","# Asumsikan model Anda sudah dilatih dan tersedia dalam variabel `model`\n","# Konversi model ke format TFLite\n","converter = tf.lite.TFLiteConverter.from_keras_model(cnn_model)\n","tflite_model = converter.convert()\n","\n","# Simpan model TFLite ke dalam file di Google Colab\n","tflite_model_path = '/content/model.tflite'  # Anda bisa mengubah path penyimpanan\n","with open(tflite_model_path, 'wb') as f:\n","    f.write(tflite_model)\n","\n","print(f\"Model berhasil dikonversi ke TFLite dan disimpan di {tflite_model_path}\")\n"],"metadata":{"colab":{"base_uri":"https://localhost:8080/"},"id":"jsrGgXTvMZkj","executionInfo":{"status":"ok","timestamp":1736932149652,"user_tz":-420,"elapsed":3265,"user":{"displayName":"Michael Emmanuel","userId":"13007318264129059752"}},"outputId":"a647a52d-9a00-4401-8720-26ce8a3927b0"},"execution_count":null,"outputs":[{"output_type":"stream","name":"stdout","text":["Saved artifact at '/tmp/tmpjeup63y2'. The following endpoints are available:\n","\n","* Endpoint 'serve'\n","  args_0 (POSITIONAL_ONLY): TensorSpec(shape=(None, 200, 200, 3), dtype=tf.float32, name='keras_tensor_5')\n","Output Type:\n","  TensorSpec(shape=(None, 1), dtype=tf.float32, name=None)\n","Captures:\n","  136508156294448: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508156292336: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508198479728: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508156419184: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508156419360: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508156415840: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508157760736: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508156418304: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508156418128: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508156418656: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508156420768: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","  136508156422000: TensorSpec(shape=(), dtype=tf.resource, name=None)\n","Model berhasil dikonversi ke TFLite dan disimpan di /content/model.tflite\n"]}]}]}