keras ConvLSTM2D 的简单应用

最新推荐文章于 2025-05-29 21:17:41 发布

原创

最新推荐文章于 2025-05-29 21:17:41 发布 · 5.1k 阅读

20 ·

CC 4.0 BY-SA版权

本文深入探讨了ConvLSTM2D层在Keras中的应用，通过实例详细解析了return_sequences与return_state参数的作用，展示了如何获取不同时间点的状态及最终状态。

示例1：返回各时刻状态

import tensorflow as tf
import numpy as np
import keras
from keras.layers import ConvLSTM2D

lstm_input = np.random.random((4,6,30,30,3)).astype(np.float32)
lstm_input = tf.convert_to_tensor(lstm_input)

lstm_out1 = ConvLSTM2D(filters=1,kernel_size=[5,5],strides=(1,1),padding='valid',activation='relu',
                       input_shape=(6,30,30,3),return_sequences=True)(lstm_input)
lstm_out2 = ConvLSTM2D(filters=2,kernel_size=[5,5],strides=(1,1),padding='valid',activation='relu',
                       return_sequences=True)(lstm_out1)
lstm_out3 = ConvLSTM2D(filters=3,kernel_size=[5,5],strides=(1,1),padding='valid',activation='relu',
                       return_sequences=True)(lstm_out2)

with tf.Session() as sess:
    sess.run(tf.global_variables_initializer())
    lstm_out1_,lstm_out2_,lstm_out3_ = sess.run([lstm_out1,lstm_out2,lstm_out3])
    print(lstm_out1_.shape)
    print(lstm_out2_.shape)
    print(lstm_out3_.sh