Description Usage Arguments Value
This layer does upsampling using a subpixel CNN, sometimes called a pixel shuffle. Essentially it shuffles together channels such that each new pixel created uses one of the channels in for that pixel in the previous layer.
1 2 | layer_subpixel_conv2d(object, scale = 2, name = NULL,
trainable = TRUE)
|
object |
A tensor input to the layer |
scale |
Upsample by this much (2 = 2x) |
name |
Optional name for the layer |
trainable |
Is the layer trainable? |
A tensor output
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.