Keras 3 API文档 / 层API / 重塑层 / UpSampling2D 层

UpSampling2D 层

[源代码]

UpSampling2D

keras.layers.UpSampling2D(
    size=(2, 2), data_format=None, interpolation="nearest", **kwargs
)

用于2D输入的上采样层。

该实现使用插值重采样,由interpolation参数指定。使用interpolation=nearest重复数据的行和列。

示例

>>> input_shape = (2, 2, 1, 3)
>>> x = np.arange(np.prod(input_shape)).reshape(input_shape)
>>> print(x)
[[[[ 0  1  2]]
  [[ 3  4  5]]]
 [[[ 6  7  8]]
  [[ 9 10 11]]]]
>>> y = keras.layers.UpSampling2D(size=(1, 2))(x)
>>> print(y)
[[[[ 0  1  2]
   [ 0  1  2]]
  [[ 3  4  5]
   [ 3  4  5]]]
 [[[ 6  7  8]
   [ 6  7  8]]
  [[ 9 10 11]
   [ 9 10 11]]]]

参数

  • size: 整数,或2个整数的元组。行和列的上采样因子。
  • data_format: 字符串,取值可以是"channels_last"(默认)或"channels_first"。输入中的维度顺序。"channels_last"对应于形状为(batch_size, height, width, channels)的输入,而"channels_first"对应于形状为(batch_size, channels, height, width)的输入。如果未指定,则使用Keras配置文件~/.keras/keras.json中的image_data_format值(如果存在),否则默认为"channels_last"。默认为"channels_last"
  • interpolation: 字符串,取值可以是"bicubic", "bilinear", "lanczos3", "lanczos5", "nearest"

输入形状

形状为 4D 张量: - 如果 data_format"channels_last"(batch_size, rows, cols, channels) - 如果 data_format"channels_first"(batch_size, channels, rows, cols)

输出形状

4D张量,形状如下:- 如果data_format"channels_last"(batch_size, upsampled_rows, upsampled_cols, channels) - 如果data_format"channels_first"(batch_size, channels, upsampled_rows, upsampled_cols)