fix Conv2DTranspose API, test=develop (#21403)

paddle_tiny_install
zhongpu 5 years ago committed by liym27
parent fbf9eca0d3
commit 8777e8c1e9

@ -2307,8 +2307,8 @@ class Conv2DTranspose(layers.Layer):
filter, and dilations, strides, paddings. Input and output filter, and dilations, strides, paddings. Input and output
are in NCHW format. Where N is batch size, C is the number of feature map, are in NCHW format. Where N is batch size, C is the number of feature map,
H is the height of the feature map, and W is the width of the feature map. H is the height of the feature map, and W is the width of the feature map.
Filter's shape is [MCHW] , where M is the number of output feature map, Filter's shape is [MCHW] , where M is the number of input feature map,
C is the number of input feature map, H is the height of the filter, C is the number of output feature map, H is the height of the filter,
and W is the width of the filter. If the groups is greater than 1, and W is the width of the filter. If the groups is greater than 1,
C will equal the number of input feature map divided by the groups. C will equal the number of input feature map divided by the groups.
If bias attribution and activation type are provided, bias is added to If bias attribution and activation type are provided, bias is added to

Loading…
Cancel
Save