You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently, the support for UpSampling layers is missing and it would be great if we could add support for these layers as well. The desired PR addressing this issue should include:
Implementation of layers classes named as UpSampling1D, UpSampling2D and UpSampling3D.
Common hierarchy of all UpSampling layers with abstract class with the common functionality
Documentation of layer and all non-private methods
JUnit tests in api module
Support for export of layers to JSON (see ModelSaver.kt)
Support for import of layers from JSON (see ModelLoader.kt)
NOTE: for the moment, there is no need to add support for "data format" (i.e. channels last vs. channels first) in your PR; you can simply assume the channels are always in the last dimension.
The text was updated successfully, but these errors were encountered:
It's a very interesting bunch of layers, @mkaze good luck with that, it includes a lot of functionality in the Keras, we should try to avoid repeating all of it, ignoring things not related to the current state of the KotlinDL
Currently, the support for UpSampling layers is missing and it would be great if we could add support for these layers as well. The desired PR addressing this issue should include:
UpSampling1D
,UpSampling2D
andUpSampling3D
.api
moduleAlso, if needed, you can take a look at Keras documentation for these layers: UpSampling1D, UpSamplling2D and UpSampling3D.
NOTE: for the moment, there is no need to add support for "data format" (i.e. channels last vs. channels first) in your PR; you can simply assume the channels are always in the last dimension.
The text was updated successfully, but these errors were encountered: