Chen Qian
Chen Qian
Sorry I don't have enough insights on this one. Will triage in the meeting.
Sorry for missing this issue (somehow it slipped away). I like the idea of flexible dtype, actually in the [experimental Keras optimizer](https://www.tensorflow.org/api_docs/python/tf/keras/optimizers/experimental/Optimizer#add_variable), there is a method `add_variable` that suits your...
@k-w-w Kathy, mind taking a look at this issue? Thanks!
@DavidLandup0 Thanks for reporting the PR! I would like to understand more about your concern - are you saying it is possible that the input does not have a feature...
@DavidLandup0 Sorry for getting back late! A little trapped on random stuff. Thanks for putting up the code example, it's very clean! We talked about this issue when we design...
@adhadse Definitely! What is the problem you see?
Got it. Basically here the purpose is to showcase how the mask generator can fit in a workflow, so we just use any model that could run. In fact, the...
@abheesht17 Thanks for opening this feature request! The idea of the paper is definitely interesting! But at this moment I am not convinced that gMLP can be a good replacement...
It's for general purpose, I am thinking we should have some components for eval purposes.