AveragePooling1D
类keras.layers.AveragePooling1D(
pool_size, strides=None, padding="valid", data_format=None, name=None, **kwargs
)
用于时序数据的平均池化。
通过对 pool_size
定义的窗口取平均值来对输入表示进行下采样。窗口会按 strides
移动。使用 "valid" 填充选项时,输出形状为:output_shape = (input_shape - pool_size + 1) / strides)
使用 "same" 填充选项时,输出形状为:output_shape = input_shape / strides
参数
pool_size
。"valid"
或 "same"
(不区分大小写)。"valid"
表示不填充。"same"
表示对输入进行均匀的左右或上下填充,使得输出具有与输入相同的宽度/高度维度。"channels_last"
或 "channels_first"
。输入中维度的顺序。"channels_last"
对应形状为 (batch, steps, features)
的输入,而 "channels_first"
对应形状为 (batch, features, steps)
的输入。它默认为 Keras 配置文件 ~/.keras/keras.json
中找到的 image_data_format
值。如果从未设置,则为 "channels_last"
。输入形状
data_format="channels_last"
:形状为 (batch_size, steps, features)
的 3D 张量。data_format="channels_first"
:形状为 (batch_size, features, steps)
的 3D 张量。输出形状
data_format="channels_last"
:形状为 (batch_size, downsampled_steps, features)
的 3D 张量。data_format="channels_first"
:形状为 (batch_size, features, downsampled_steps)
的 3D 张量。示例
strides=1
且 padding="valid"
>>> x = np.array([1., 2., 3., 4., 5.])
>>> x = np.reshape(x, [1, 5, 1])
>>> avg_pool_1d = keras.layers.AveragePooling1D(pool_size=2,
... strides=1, padding="valid")
>>> avg_pool_1d(x)
strides=2
且 padding="valid"
>>> x = np.array([1., 2., 3., 4., 5.])
>>> x = np.reshape(x, [1, 5, 1])
>>> avg_pool_1d = keras.layers.AveragePooling1D(pool_size=2,
... strides=2, padding="valid")
>>> avg_pool_1d(x)
strides=1
且 padding="same"
>>> x = np.array([1., 2., 3., 4., 5.])
>>> x = np.reshape(x, [1, 5, 1])
>>> avg_pool_1d = keras.layers.AveragePooling1D(pool_size=2,
... strides=1, padding="same")
>>> avg_pool_1d(x)