A modern programming language that makes developers happier.

kotlinlang

Hi folks, great to be here. :wave::skin-tone-2:
A general question: I'm currently working on <https://github.com/JetBrains/KotlinDL/issues/50>
As we all know there is a lot of functionality missing from the tensorflow API compared to the Python version.
How do we feel about introducing convenience functions that are missing?

For example: The `Identity Initializer` is basically:
`tf.eye(*shape, dtype)` in Keras.
Buuuuuut: We have no `tf.eye`
So: do we need stuff like this? And if yes: Where do we put it?

<@U023NEP0PA8> first of all, I happy to see your PR and your question here. I see that you implemented it without tf.eye usage as a composition of another available tf ops. There are two ways: doing nothing about tf.eye or create an extension or util function which do the same (but built on a few tf ops). I never extract the same pieces of code as tf extensions, but faced with the same issue in a few moments around gradient clipping, or making fake concat op via padding. So you could be a first contributor here. If you need a separate issue and separate PR for that for tracking it separately feel free to do it. <https://github.com/JetBrains/KotlinDL/blob/master/api/src/main/kotlin/org/jetbrains/kotlinx/dl/api/core/util/TF.kt|There are> some analogues of Keras functions (used as an analogue of tf.moments) and <https://github.com/JetBrains/KotlinDL/blob/master/api/src/main/kotlin/org/jetbrains/kotlinx/dl/api/extension/TensorExtensionFunctions.kt|here> you could find some useful extension functions for the TF Tensor. You could add your implementation here, in "extension" package

Thanks a lot for the input Alexey. Much appreciated!