deepali0914/image_classifier

1{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.12.13","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[],"dockerImageVersionId":28755,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"import os, warnings\nimport matplotlib.pyplot as plt\nfrom matplotlib import gridspec\nimport numpy as np\nimport tensorflow as tf\nfrom tensorflow.keras import layers\nfrom tensorflow.keras.preprocessing import image_dataset_from_directory\n\n# Reproducibility\ndef set_seed(seed=31415):\n    np.random.seed(seed)\n    tf.random.set_seed(seed)\n    os.environ['PYTHONHASHSEED'] = str(seed)\nset_seed()\n\n# Matplotlib defaults\nplt.rc('figure', autolayout=True)\nplt.rc('axes', labelweight='bold', labelsize='large',\n       titleweight='bold', titlesize=18, titlepad=10)\nplt.rc('image', cmap='magma')\nwarnings.filterwarnings(\"ignore\")\n\nAUTOTUNE = tf.data.experimental.AUTOTUNE\n\nfor dirpath, dirnames, filenames in os.walk('/kaggle/input'):\n    print(dirpath, len(filenames), 'files')\n\nbase_path = '/kaggle/input/datasets/rahmasleam/intel-image-dataset/Intel Image Dataset'\nclass_names = os.listdir(base_path)\nprint(class_names)\n\nimport pandas as pd\n\ndata = []\n\nfor class_name in class_names:\n    class_folder = os.path.join(base_path, class_name)\n    files = os.listdir(class_folder)\n    for filename in files:\n        filepath = os.path.join(class_folder, filename)\n        data.append({'filepath': filepath, 'label': class_name})\n\ndf = pd.DataFrame(data)\n\nfrom sklearn.model_selection import train_test_split\n\nX_train, X_temp, y_train, y_temp = train_test_split(\n    df['filepath'], df['label'], \n    test_size=0.3, \n    stratify=df['label'],  # <-- This guarantees proportional splitting\n    random_state=42\n)\n\nX_val, X_test, y_val, y_test = train_test_split(\n    X_temp, y_temp, \n    test_size=0.5, \n    stratify=y_temp,  # <-- This guarantees proportional splitting\n    random_state=42\n)\n\nprint(len(X_train), len(X_val), len(X_test))\nprint(y_train.value_counts(normalize=True))\nprint(y_val.value_counts(normalize=True))\nprint(y_test.value_counts(normalize=True))\n\nclass_names = sorted(df['label'].unique())\nlabel_to_index = {name: i for i, name in enumerate(class_names)}\nprint(label_to_index)\n\ny_train_encoded = y_train.map(label_to_index).values\ny_val_encoded = y_val.map(label_to_index).values\ny_test_encoded = y_test.map(label_to_index).values\n\nIMG_SIZE = 150\n\ndef load_and_preprocess_image(filepath, label):\n    image = tf.io.read_file(filepath)               # read the raw file bytes from disk\n    image = tf.image.decode_jpeg(image, channels=3)  # turn those bytes into actual pixel values\n    image = tf.image.resize(image, [IMG_SIZE, IMG_SIZE])  # force it to be exactly 150x150\n    image = tf.cast(image, tf.float32) / 255.0        # scale pixel values from 0-255 down to 0-1\n    return image, label\n\nBATCH_SIZE = 32\nAUTOTUNE = tf.data.experimental.AUTOTUNE  # you may already have this from your boilerplate\n\ndef make_dataset(filepaths, labels, shuffle=False):\n    ds = tf.data.Dataset.from_tensor_slices((filepaths.values, labels))\n    ds = ds.map(load_and_preprocess_image, num_parallel_calls=AUTOTUNE)\n    if shuffle:\n        ds = ds.shuffle(buffer_size=len(filepaths))\n    ds = ds.batch(BATCH_SIZE)\n    ds = ds.prefetch(buffer_size=AUTOTUNE)\n    return ds\n\ntrain_ds = make_dataset(X_train, y_train_encoded, shuffle=True)\nval_ds = make_dataset(X_val, y_val_encoded, shuffle=False)\ntest_ds = make_dataset(X_test, y_test_encoded, shuffle=False)\n\nfrom tensorflow import keras\nfrom tensorflow.keras import layers\n\ndata_augmentation = keras.Sequential([\n    layers.RandomFlip(\"horizontal\"),\n    layers.RandomRotation(0.1),\n    layers.RandomZoom(0.1),\n])\n\nmodel = keras.Sequential([\n    data_augmentation,\n    \n    # Block One\n    layers.Conv2D(filters=32, kernel_size=3, padding='same',\n                  input_shape=[150, 150, 3]),\n    layers.BatchNormalization(),\n    layers.Activation('relu'),\n    layers.MaxPool2D(),\n\n    # Block Two\n    layers.Conv2D(filters=64, kernel_size=3, padding='same'),\n    layers.BatchNormalization(),\n    layers.Activation('relu'),\n    layers.MaxPool2D(),\n\n    #Block Three\n    layers.Conv2D(filters=128, kernel_size=3, padding='same'),\n    layers.BatchNormalization(),\n    layers.Activation('relu'),\n    layers.MaxPool2D(),\n\n    # Head\n    layers.Flatten(),\n    layers.Dense(128, activation='relu'),\n    layers.Dropout(0.2),\n    layers.Dense(6, activation='softmax'),\n])\n\nmodel.compile(\n    optimizer='adam',\n    loss='sparse_categorical_crossentropy',\n    metrics=['accuracy']\n)\n\nhistory = model.fit(\n    train_ds,\n    validation_data=val_ds,\n    epochs=50,\n)\n\nhistory_frame = pd.DataFrame(history.history)\nhistory_frame.loc[:, ['loss', 'val_loss']].plot()\nhistory_frame.loc[:, ['accuracy', 'val_accuracy']].plot();\n\ntrain_loss, train_accuracy = model.evaluate(train_ds)\nval_loss, val_accuracy = model.evaluate(val_ds)\nprint(f\"Train — loss: {train_loss:.4f}, accuracy: {train_accuracy:.4f}\")\nprint(f\"Val   — loss: {val_loss:.4f}, accuracy: {val_accuracy:.4f}\")","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true,"execution":{"iopub.status.busy":"2026-07-24T20:21:01.209227Z","iopub.execute_input":"2026-07-24T20:21:01.209699Z","iopub.status.idle":"2026-07-24T20:24:34.924376Z","shell.execute_reply.started":"2026-07-24T20:21:01.209668Z","shell.execute_reply":"2026-07-24T20:24:34.923529Z"}},"outputs":[{"name":"stdout","text":"/kaggle/input 0 files\n/kaggle/input/datasets 0 files\n/kaggle/input/datasets/rahmasleam 0 files\n/kaggle/input/datasets/rahmasleam/intel-image-dataset 0 files\n/kaggle/input/datasets/rahmasleam/intel-image-dataset/Intel Image Dataset 0 files\n/kaggle/input/datasets/rahmasleam/intel-image-dataset/Intel Image Dataset/mountain 525 files\n/kaggle/input/datasets/rahmasleam/intel-image-dataset/Intel Image Dataset/street 501 files\n/kaggle/input/datasets/rahmasleam/intel-image-dataset/Intel Image Dataset/buildings 437 files\n/kaggle/input/datasets/rahmasleam/intel-image-dataset/Intel Image Dataset/sea 510 files\n/kaggle/input/datasets/rahmasleam/intel-image-dataset/Intel Image Dataset/forest 474 files\n/kaggle/input/datasets/rahmasleam/intel-image-dataset/Intel Image Dataset/glacier 553 files\n['mountain', 'street', 'buildings', 'sea', 'forest', 'glacier']\n2100 450 450\nlabel\nglacier      0.184286\nmountain     0.174762\nsea          0.170000\nstreet       0.167143\nforest       0.158095\nbuildings    0.145714\nName: proportion, dtype: float64\nlabel\nglacier      0.184444\nmountain     0.175556\nsea          0.171111\nstreet       0.166667\nforest       0.157778\nbuildings    0.144444\nName: proportion, dtype: float64\nlabel\nglacier      0.184444\nmountain     0.175556\nsea          0.168889\nstreet       0.166667\nforest       0.157778\nbuildings    0.146667\nName: proportion, dtype: float64\n{'buildings': 0, 'forest': 1, 'glacier': 2, 'mountain': 3, 'sea': 4, 'street': 5}\n","output_type":"stream"},{"name":"stderr","text":"WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\nI0000 00:00:1784924480.565414      58 gpu_device.cc:2020] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 15511 MB memory:  -> device: 0, name: Tesla P100-PCIE-16GB, pci bus id: 0000:00:04.0, compute capability: 6.0\n","output_type":"stream"},{"name":"stdout","text":"Epoch 1/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m12s\u001b[0m 54ms/step - accuracy: 0.2843 - loss: 5.5266 - val_accuracy: 0.1978 - val_loss: 1.7465\nEpoch 2/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.3071 - loss: 1.6332 - val_accuracy: 0.1711 - val_loss: 1.8701\nEpoch 3/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.3305 - loss: 1.5813 - val_accuracy: 0.2244 - val_loss: 1.7248\nEpoch 4/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.3657 - loss: 1.5158 - val_accuracy: 0.2222 - val_loss: 1.6656\nEpoch 5/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 43ms/step - accuracy: 0.3948 - loss: 1.4575 - val_accuracy: 0.2067 - val_loss: 1.9942\nEpoch 6/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.3933 - loss: 1.4293 - val_accuracy: 0.4644 - val_loss: 1.3183\nEpoch 7/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.3986 - loss: 1.3737 - val_accuracy: 0.4444 - val_loss: 1.2367\nEpoch 8/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4110 - loss: 1.3460 - val_accuracy: 0.2467 - val_loss: 2.7070\nEpoch 9/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4033 - loss: 1.3277 - val_accuracy: 0.4533 - val_loss: 1.2400\nEpoch 10/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.3657 - loss: 1.3336 - val_accuracy: 0.4867 - val_loss: 1.1777\nEpoch 11/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4124 - loss: 1.3222 - val_accuracy: 0.4689 - val_loss: 1.3219\nEpoch 12/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4157 - loss: 1.2976 - val_accuracy: 0.3511 - val_loss: 1.5607\nEpoch 13/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4300 - loss: 1.3129 - val_accuracy: 0.3044 - val_loss: 1.8560\nEpoch 14/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4343 - loss: 1.2868 - val_accuracy: 0.4800 - val_loss: 1.1556\nEpoch 15/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4305 - loss: 1.2786 - val_accuracy: 0.4978 - val_loss: 1.2317\nEpoch 16/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4771 - loss: 1.2323 - val_accuracy: 0.4911 - val_loss: 1.2305\nEpoch 17/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4605 - loss: 1.2248 - val_accuracy: 0.4600 - val_loss: 1.2719\nEpoch 18/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4467 - loss: 1.2486 - val_accuracy: 0.5156 - val_loss: 1.1764\nEpoch 19/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4486 - loss: 1.2368 - val_accuracy: 0.4778 - val_loss: 1.1656\nEpoch 20/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4700 - loss: 1.2088 - val_accuracy: 0.4800 - val_loss: 1.1660\nEpoch 21/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4776 - loss: 1.2115 - val_accuracy: 0.4778 - val_loss: 1.1994\nEpoch 22/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m3s\u001b[0m 43ms/step - accuracy: 0.4767 - loss: 1.2115 - val_accuracy: 0.4600 - val_loss: 1.3055\nEpoch 23/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4924 - loss: 1.1624 - val_accuracy: 0.4356 - val_loss: 1.3188\nEpoch 24/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4971 - loss: 1.1770 - val_accuracy: 0.5333 - val_loss: 1.1645\nEpoch 25/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 42ms/step - accuracy: 0.4848 - loss: 1.1743 - val_accuracy: 0.4667 - val_loss: 1.1551\nEpoch 26/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 42ms/step - accuracy: 0.4738 - loss: 1.1889 - val_accuracy: 0.5200 - val_loss: 1.2200\nEpoch 27/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4957 - loss: 1.2236 - val_accuracy: 0.4822 - val_loss: 1.1312\nEpoch 28/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 42ms/step - accuracy: 0.5010 - loss: 1.1487 - val_accuracy: 0.5311 - val_loss: 1.0664\nEpoch 29/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.4990 - loss: 1.1527 - val_accuracy: 0.5178 - val_loss: 1.1189\nEpoch 30/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 44ms/step - accuracy: 0.5090 - loss: 1.1785 - val_accuracy: 0.5822 - val_loss: 1.1351\nEpoch 31/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5133 - loss: 1.1220 - val_accuracy: 0.4889 - val_loss: 1.1789\nEpoch 32/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5067 - loss: 1.1535 - val_accuracy: 0.5622 - val_loss: 1.1781\nEpoch 33/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5367 - loss: 1.1229 - val_accuracy: 0.3556 - val_loss: 1.7544\nEpoch 34/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5114 - loss: 1.1407 - val_accuracy: 0.5378 - val_loss: 1.1268\nEpoch 35/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 42ms/step - accuracy: 0.5219 - loss: 1.1172 - val_accuracy: 0.5222 - val_loss: 1.1608\nEpoch 36/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 42ms/step - accuracy: 0.5348 - loss: 1.1045 - val_accuracy: 0.5844 - val_loss: 1.0579\nEpoch 37/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 44ms/step - accuracy: 0.5086 - loss: 1.1077 - val_accuracy: 0.5467 - val_loss: 1.0055\nEpoch 38/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5176 - loss: 1.1291 - val_accuracy: 0.5333 - val_loss: 1.1037\nEpoch 39/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5186 - loss: 1.1321 - val_accuracy: 0.5311 - val_loss: 1.1433\nEpoch 40/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5357 - loss: 1.0891 - val_accuracy: 0.6200 - val_loss: 1.0383\nEpoch 41/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5024 - loss: 1.1585 - val_accuracy: 0.5044 - val_loss: 1.1910\nEpoch 42/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5119 - loss: 1.1103 - val_accuracy: 0.4933 - val_loss: 1.0288\nEpoch 43/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5262 - loss: 1.0877 - val_accuracy: 0.5467 - val_loss: 1.1505\nEpoch 44/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5333 - loss: 1.1080 - val_accuracy: 0.5933 - val_loss: 1.0368\nEpoch 45/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5405 - loss: 1.0655 - val_accuracy: 0.6178 - val_loss: 0.9622\nEpoch 46/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 42ms/step - accuracy: 0.5500 - loss: 1.0775 - val_accuracy: 0.5244 - val_loss: 1.2238\nEpoch 47/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5543 - loss: 1.0613 - val_accuracy: 0.5844 - val_loss: 0.9794\nEpoch 48/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5576 - loss: 1.0624 - val_accuracy: 0.5889 - val_loss: 1.0473\nEpoch 49/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 43ms/step - accuracy: 0.5438 - loss: 1.0465 - val_accuracy: 0.5733 - val_loss: 0.9836\nEpoch 50/50\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m4s\u001b[0m 42ms/step - accuracy: 0.5638 - loss: 1.0396 - val_accuracy: 0.5800 - val_loss: 0.9797\n\u001b[1m66/66\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m2s\u001b[0m 11ms/step - accuracy: 0.6362 - loss: 0.8505\n\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 12ms/step - accuracy: 0.5800 - loss: 0.9797\nTrain — loss: 0.8505, accuracy: 0.6362\nVal   — loss: 0.9797, accuracy: 0.5800\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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\n"},"metadata":{}},{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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\n"},"metadata":{}}],"execution_count":1},{"cell_type":"code","source":"model.save('scene_classifier.keras')\n!pip install gradio","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-07-24T20:31:37.094443Z","iopub.execute_input":"2026-07-24T20:31:37.095158Z","iopub.status.idle":"2026-07-24T20:31:41.596031Z","shell.execute_reply.started":"2026-07-24T20:31:37.095129Z","shell.execute_reply":"2026-07-24T20:31:41.595294Z"}},"outputs":[{"name":"stdout","text":"Requirement already satisfied: gradio in /usr/local/lib/python3.12/dist-packages (5.50.0)\nRequirement already satisfied: aiofiles<25.0,>=22.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (22.1.0)\nRequirement already satisfied: anyio<5.0,>=3.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (4.13.0)\nRequirement already satisfied: brotli>=1.1.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (1.2.0)\nRequirement already satisfied: fastapi<1.0,>=0.115.2 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.136.1)\nRequirement already satisfied: ffmpy in /usr/local/lib/python3.12/dist-packages (from gradio) (1.0.0)\nRequirement already satisfied: gradio-client==1.14.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (1.14.0)\nRequirement already satisfied: groovy~=0.1 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.1.2)\nRequirement already satisfied: httpx<1.0,>=0.24.1 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.28.1)\nRequirement already satisfied: huggingface-hub<2.0,>=0.33.5 in /usr/local/lib/python3.12/dist-packages (from gradio) (1.11.0)\nRequirement already satisfied: jinja2<4.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (3.1.6)\nRequirement already satisfied: markupsafe<4.0,>=2.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (3.0.3)\nRequirement already satisfied: numpy<3.0,>=1.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (2.0.2)\nRequirement already satisfied: orjson~=3.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (3.11.8)\nRequirement already satisfied: packaging in /usr/local/lib/python3.12/dist-packages (from gradio) (26.1)\nRequirement already satisfied: pandas<3.0,>=1.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (2.3.3)\nRequirement already satisfied: pillow<12.0,>=8.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (11.3.0)\nRequirement already satisfied: pydantic<=2.12.3,>=2.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (2.12.3)\nRequirement already satisfied: pydub in /usr/local/lib/python3.12/dist-packages (from gradio) (0.25.1)\nRequirement already satisfied: python-multipart>=0.0.18 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.0.26)\nRequirement already satisfied: pyyaml<7.0,>=5.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (6.0.3)\nRequirement already satisfied: ruff>=0.9.3 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.15.11)\nRequirement already satisfied: safehttpx<0.2.0,>=0.1.6 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.1.7)\nRequirement already satisfied: semantic-version~=2.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (2.10.0)\nRequirement already satisfied: starlette<1.0,>=0.40.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.52.1)\nRequirement already satisfied: tomlkit<0.14.0,>=0.12.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.13.3)\nRequirement already satisfied: typer<1.0,>=0.12 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.24.2)\nRequirement already satisfied: typing-extensions~=4.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (4.15.0)\nRequirement already satisfied: uvicorn>=0.14.0 in /usr/local/lib/python3.12/dist-packages (from gradio) (0.46.0)\nRequirement already satisfied: fsspec in /usr/local/lib/python3.12/dist-packages (from gradio-client==1.14.0->gradio) (2025.3.0)\nRequirement already satisfied: websockets<16.0,>=13.0 in /usr/local/lib/python3.12/dist-packages (from gradio-client==1.14.0->gradio) (15.0.1)\nRequirement already satisfied: idna>=2.8 in /usr/local/lib/python3.12/dist-packages (from anyio<5.0,>=3.0->gradio) (3.13)\nRequirement already satisfied: typing-inspection>=0.4.2 in /usr/local/lib/python3.12/dist-packages (from fastapi<1.0,>=0.115.2->gradio) (0.4.2)\nRequirement already satisfied: annotated-doc>=0.0.2 in /usr/local/lib/python3.12/dist-packages (from fastapi<1.0,>=0.115.2->gradio) (0.0.4)\nRequirement already satisfied: certifi in /usr/local/lib/python3.12/dist-packages (from httpx<1.0,>=0.24.1->gradio) (2026.4.22)\nRequirement already satisfied: httpcore==1.* in /usr/local/lib/python3.12/dist-packages (from httpx<1.0,>=0.24.1->gradio) (1.0.9)\nRequirement already satisfied: h11>=0.16 in /usr/local/lib/python3.12/dist-packages (from httpcore==1.*->httpx<1.0,>=0.24.1->gradio) (0.16.0)\nRequirement already satisfied: filelock>=3.10.0 in /usr/local/lib/python3.12/dist-packages (from huggingface-hub<2.0,>=0.33.5->gradio) (3.29.0)\nRequirement already satisfied: hf-xet<2.0.0,>=1.4.3 in /usr/local/lib/python3.12/dist-packages (from huggingface-hub<2.0,>=0.33.5->gradio) (1.4.3)\nRequirement already satisfied: tqdm>=4.42.1 in /usr/local/lib/python3.12/dist-packages (from huggingface-hub<2.0,>=0.33.5->gradio) (4.67.3)\nRequirement already satisfied: python-dateutil>=2.8.2 in /usr/local/lib/python3.12/dist-packages (from pandas<3.0,>=1.0->gradio) (2.9.0.post0)\nRequirement already satisfied: pytz>=2020.1 in /usr/local/lib/python3.12/dist-packages (from pandas<3.0,>=1.0->gradio) (2025.2)\nRequirement already satisfied: tzdata>=2022.7 in /usr/local/lib/python3.12/dist-packages (from pandas<3.0,>=1.0->gradio) (2026.1)\nRequirement already satisfied: annotated-types>=0.6.0 in /usr/local/lib/python3.12/dist-packages (from pydantic<=2.12.3,>=2.0->gradio) (0.7.0)\nRequirement already satisfied: pydantic-core==2.41.4 in /usr/local/lib/python3.12/dist-packages (from pydantic<=2.12.3,>=2.0->gradio) (2.41.4)\nRequirement already satisfied: click>=8.2.1 in /usr/local/lib/python3.12/dist-packages (from typer<1.0,>=0.12->gradio) (8.3.3)\nRequirement already satisfied: shellingham>=1.3.0 in /usr/local/lib/python3.12/dist-packages (from typer<1.0,>=0.12->gradio) (1.5.4)\nRequirement already satisfied: rich>=12.3.0 in /usr/local/lib/python3.12/dist-packages (from typer<1.0,>=0.12->gradio) (13.9.4)\nRequirement already satisfied: six>=1.5 in /usr/local/lib/python3.12/dist-packages (from python-dateutil>=2.8.2->pandas<3.0,>=1.0->gradio) (1.17.0)\nRequirement already satisfied: markdown-it-py>=2.2.0 in /usr/local/lib/python3.12/dist-packages (from rich>=12.3.0->typer<1.0,>=0.12->gradio) (4.0.0)\nRequirement already satisfied: pygments<3.0.0,>=2.13.0 in /usr/local/lib/python3.12/dist-packages (from rich>=12.3.0->typer<1.0,>=0.12->gradio) (2.20.0)\nRequirement already satisfied: mdurl~=0.1 in /usr/local/lib/python3.12/dist-packages (from markdown-it-py>=2.2.0->rich>=12.3.0->typer<1.0,>=0.12->gradio) (0.1.2)\n","output_type":"stream"}],"execution_count":2},{"cell_type":"code","source":"import gradio as gr\nimport numpy as np\nfrom tensorflow import keras\n\nmodel = keras.models.load_model('scene_classifier.keras')\nclass_names = ['buildings', 'forest', 'glacier', 'mountain', 'sea', 'street']  # must match your label_to_index order\n\ndef predict_image(img):\n    img = img.resize((150, 150))                    # match your training input size\n    img_array = np.array(img) / 255.0                # normalize, same as training\n    img_array = np.expand_dims(img_array, axis=0)    # add batch dimension: (150,150,3) -> (1,150,150,3)\n    predictions = model.predict(img_array)[0]         # get prediction probabilities for the 6 classes\n    return {class_names[i]: float(predictions[i]) for i in range(len(class_names))}","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-07-24T20:31:46.572730Z","iopub.execute_input":"2026-07-24T20:31:46.573014Z","iopub.status.idle":"2026-07-24T20:31:57.764429Z","shell.execute_reply.started":"2026-07-24T20:31:46.572988Z","shell.execute_reply":"2026-07-24T20:31:57.763749Z"}},"outputs":[],"execution_count":3},{"cell_type":"code","source":"demo = gr.Interface(\n    fn=predict_image,\n    inputs=gr.Image(type='pil'),\n    outputs=gr.Label(num_top_classes=6),\n    title=\"Natural Scene Classifier\",\n    description=\"Upload a photo of a natural scene (buildings, forest, glacier, mountain, sea, or street) and see the model's prediction.\"\n)\n\ndemo.launch()","metadata":{"trusted":true,"execution":{"iopub.status.busy":"2026-07-24T20:32:03.668014Z","iopub.execute_input":"2026-07-24T20:32:03.668768Z","iopub.status.idle":"2026-07-24T20:32:04.994374Z","shell.execute_reply.started":"2026-07-24T20:32:03.668741Z","shell.execute_reply":"2026-07-24T20:32:04.993813Z"}},"outputs":[{"name":"stdout","text":"* Running on local URL:  http://127.0.0.1:7860\nIt looks like you are running Gradio on a hosted Jupyter notebook, which requires `share=True`. Automatically setting `share=True` (you can turn this off by setting `share=False` in `launch()` explicitly).\n\n* Running on public URL: https://a82bcc57dd3d56e426.gradio.live\n\nThis share link expires in 1 week. For free permanent hosting and GPU upgrades, run `gradio deploy` from the terminal in the working directory to deploy to Hugging Face Spaces (https://huggingface.co/spaces)\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<IPython.core.display.HTML object>","text/html":"<div><iframe src=\"https://a82bcc57dd3d56e426.gradio.live\" width=\"100%\" height=\"500\" allow=\"autoplay; camera; microphone; clipboard-read; clipboard-write;\" frameborder=\"0\" allowfullscreen></iframe></div>"},"metadata":{}},{"execution_count":4,"output_type":"execute_result","data":{"text/plain":""},"metadata":{}},{"name":"stdout","text":"\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 238ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 31ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 29ms/step\n\u001b[1m1/1\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 32ms/step\n","output_type":"stream"}],"execution_count":4}]}