UpSampling2D 层
UpSampling2D class
keras.layers.UpSampling2D(
size=(2, 2), data_format=None, interpolation="nearest", **kwargs
)
二维输入的上采样层.
该实现使用插值调整大小,给定调整方法（由 interpolation 参数指定）.使用 interpolation=nearest 来重复数据的行和列.
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,或两个整数的元组.
行和列的上采样因子.
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" 对应输入形状
(batch_size, height, width, channels),而 "channels_first"
对应输入形状
(batch_size, channels, height, width).
如果未指定,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".
interpolation: 字符串,可以是 "bicubic"、"bilinear"、"lanczos3"、
"lanczos5"、"nearest".
输入形状:
形状为以下之一的 4D 张量:
- 如果 data_format 是 "channels_last":
(batch_size, rows, cols, channels)
- 如果 data_format 是 "channels_first":
(batch_size, channels, rows, cols)
输出形状:
形状为以下之一的 4D 张量:
- 如果 data_format 是 "channels_last":
(batch_size, upsampled_rows, upsampled_cols, channels)
- 如果 data_format 是 "channels_first":
(batch_size, channels, upsampled_rows, upsampled_cols)