diff --git a/scenery_train.py b/scenery_train.py
new file mode 100644
index 0000000..9200f7c
--- /dev/null
+++ b/scenery_train.py
@@ -0,0 +1,79 @@
+import os
+import numpy as np
+from PIL import Image
+from tensorflow.keras.preprocessing.image import ImageDataGenerator
+from tensorflow.keras.models import Sequential
+from tensorflow.keras.layers import Conv2D, MaxPooling2D, Flatten, Dense
+import tensorflow as tf
+from tensorflow.keras.callbacks import Callback, EarlyStopping
+
+# 数据路径
+train_data_dir = 'dataset/scenery/seg_train'
+test_data_dir = 'dataset/scenery/seg_test'
+batch_size = 32
+
+# 图片生成器
+train_datagen = ImageDataGenerator(rescale=1./255)
+test_datagen = ImageDataGenerator(rescale=1./255)
+
+train_generator = train_datagen.flow_from_directory(
+    train_data_dir,
+    target_size=(180, 180),
+    batch_size=batch_size,
+    class_mode='categorical'
+)
+
+test_generator = test_datagen.flow_from_directory(
+    test_data_dir,
+    target_size=(180, 180),
+    batch_size=batch_size,
+    class_mode='categorical'
+)
+
+# 配置GPU加速
+strategy = tf.distribute.MirroredStrategy()
+print('Number of devices: {}'.format(strategy.num_replicas_in_sync))
+
+# 配置早停
+early_stopping = EarlyStopping(monitor='val_loss', patience=3, verbose=1, restore_best_weights=True)
+
+with strategy.scope():
+    model = Sequential([
+        Conv2D(32, (3, 3), activation='relu', input_shape=(180, 180, 3)),
+        MaxPooling2D((2, 2)),
+        Conv2D(64, (3, 3), activation='relu'),
+        MaxPooling2D((2, 2)),
+        Conv2D(128, (3, 3), activation='relu'),
+        MaxPooling2D((2, 2)),
+        Flatten(),
+        Dense(128, activation='relu'),
+        Dense(len(train_generator.class_indices), activation='softmax')
+    ])
+    # 编译模型
+    optimizer = tf.keras.optimizers.Adam(learning_rate=0.001)
+    model.compile(optimizer=optimizer, loss='categorical_crossentropy', metrics=['accuracy'])
+
+# # 配置模型检查点，保存最优模型
+# checkpoint_path = "./model/animal_model.h5"
+# model_checkpoint = ModelCheckpoint(checkpoint_path, monitor='val_loss',
+#                                      save_best_only=True, save_weights_only=False, verbose=1)
+
+# 训练模型
+model.fit(train_generator, epochs=10, validation_data=test_generator, callbacks=[early_stopping])
+
+# 保存模型为 .h5 文件
+model.save("./model/scenery_model.h5", save_format='h5')
+
+
+#     model.compile(optimizer='adam', loss='categorical_crossentropy', metrics=['accuracy'])
+#
+# # 自定义回调函数，保存模型为 .h5 格式
+# class CustomModelCheckpoint(Callback):
+#     def on_epoch_end(self, epoch, logs=None):
+#         self.model.save("./model/scenery_model.h5")
+#
+# # 创建自定义回调函数实例
+# custom_checkpoint = CustomModelCheckpoint()
+#
+# # 训练模型时使用自定义回调函数
+# model.fit(train_generator, epochs=10, validation_data=test_generator, callbacks=[custom_checkpoint])