Cropping3D 类keras.layers.Cropping3D(
cropping=((1, 1), (1, 1), (1, 1)), data_format=None, **kwargs
)
用于 3D 数据(例如空间或时空)的裁剪层。
示例
>>> input_shape = (2, 28, 28, 10, 3)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> y = keras.layers.Cropping3D(cropping=(2, 4, 2))(x)
>>> y.shape
(2, 24, 20, 6, 3)
参数
(symmetric_dim1_crop, symmetric_dim2_crop, symmetric_dim3_crop)。((left_dim1_crop, right_dim1_crop), (left_dim2_crop, right_dim2_crop), (left_dim3_crop, right_dim3_crop))。"channels_last"(默认)或 "channels_first"。输入中维度的顺序。"channels_last" 对应形状为 (batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels) 的输入,而 "channels_first" 对应形状为 (batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3) 的输入。未指定时,使用 Keras 配置文件 ~/.keras/keras.json(如果存在)中的 image_data_format 值。默认为 "channels_last"。输入形状
形状为 5D 张量: - 如果 data_format 为 "channels_last":(batch_size, first_axis_to_crop, second_axis_to_crop, third_axis_to_crop, channels) - 如果 data_format 为 "channels_first":(batch_size, channels, first_axis_to_crop, second_axis_to_crop, third_axis_to_crop)
输出形状
形状为 5D 张量: - 如果 data_format 为 "channels_last":(batch_size, first_cropped_axis, second_cropped_axis, third_cropped_axis, channels) - 如果 data_format 为 "channels_first":(batch_size, channels, first_cropped_axis, second_cropped_axis, third_cropped_axis)