三维上采样层
UpSampling3D class
keras.layers.UpSampling3D(size=(2, 2, 2), data_format=None, **kwargs)
三维输入的上采样层.
将数据的第1、2和3维度分别重复size[0]、size[1]和size[2]次.
示例:
>>> input_shape = (2, 1, 2, 1, 3)
>>> x = np.ones(input_shape)
>>> y = keras.layers.UpSampling3D(size=(2, 2, 2))(x)
>>> y.shape
(2, 2, 4, 2, 3)
参数:
size: 整数,或3个整数的元组.
用于dim1、dim2和dim3的上采样因子.
data_format: 字符串,
可以是"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,则使用在您的Keras配置文件中找到的
image_data_format值,路径为
~/.keras/keras.json（如果存在）,否则为"channels_last".
默认为"channels_last".
输入形状:
形状为以下之一的5D张量:
- 如果data_format是"channels_last":
(batch_size, dim1, dim2, dim3, channels)
- 如果data_format是"channels_first":
(batch_size, channels, dim1, dim2, dim3)
输出形状:
形状为以下之一的5D张量:
- 如果data_format是"channels_last":
(batch_size, upsampled_dim1, upsampled_dim2, upsampled_dim3,
channels)
- 如果data_format是"channels_first":
(batch_size, channels, upsampled_dim1, upsampled_dim2,
upsampled_dim3)