三维剪裁层
Cropping3D class
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:
cropping: 整数,或3个整数的元组,或3个2个整数的元组的元组.
- 如果为整数: 相同的对称裁剪应用于深度、高度和宽度.
- 如果为3个整数的元组: 解释为深度、高度和宽度的三个不同的对称裁剪值:
(symmetric_dim1_crop, symmetric_dim2_crop, symmetric_dim3_crop).
- 如果为3个2个整数的元组的元组: 解释为
((left_dim1_crop, right_dim1_crop), (left_dim2_crop,
right_dim2_crop), (left_dim3_crop, right_dim3_crop)).
data_format: 字符串,取值为"channels_last"（默认）或
"channels_first".输入中维度的顺序.
"channels_last"对应于形状为
(batch_size, spatial_dim1, spatial_dim2, spatial_dim3, channels)的输入
而"channels_first"对应于形状为
(batch_size, channels, spatial_dim1, spatial_dim2, spatial_dim3)的输入.
如果未指定,使用在您的Keras配置文件~/.keras/keras.json中找到的image_data_format值（如果存在）.默认为
"channels_last".
输入形状:
5D张量,形状为:
- 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:
5D张量,形状为:
- 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)