Add Non-linear activation functions to PyTorch Frontend

[djl11]

Add Non-linear activation functions to PyTorch frontend:

_

Please keep in mind that the proper way to link an issue to this list is to comment "- [ ] #issue_number" while the issue's title only includes the name of the function you've chosen.

_

• threshold #4039
• threshold_ #4040
• relu #6552
• relu_ #5797
• hardtanh #4651
• hardtanh_ #4652
• hardswish #4650
• relu6 #4083
• elu #4084
• elu_ #4085
• selu #4105
• celu #4106
• leaky_relu #2723
• leaky_relu_ #4491
• prelu #4219
• rrelu #4220
• rrelu_ #4221
• glu #4397
• gelu #2981
• logsigmoid #3889
• hardshrink #4402
• tanhshrink #4492
• softsign #4401
• softplus #6537
• softmin #3966
• softmax #2725
• softshrink #4403
• gumbel_softmax
• log_softmax #4494
• tanh
• sigmoid #2480
• hardsigmoid #4263
• silu #4398
• mish #7132
• batch_norm #15814
• group_norm
• instance_norm
• layer_norm #17794
• local_response_norm #17842
• normalize #4823

[leonjovanovic]

Is #14660 open to work on or is it taken?

[AnnaTz]

Is #14660 open to work on or is it taken?

I think you can link your PR to the issue and work on it. Whoever opened it ended up adding another function instead of tanh.

[leonjovanovic]

I think you can link your PR to the issue and work on it. Whoever opened it ended up adding another function instead of tanh.

I think tanh is already done. Link

I ll do gumbel_softmax PR

[faiqali1]

All the functions are complete.
Mark this as closed?

[AnnaTz]

All the functions are complete. Mark this as closed?

There is one function remaining.

[faiqali1]

#17586
gumbel_softmax has been merged?
any other issues left?

[AnnaTz]

#17586 gumbel_softmax has been merged? any other issues left?

Oh it wasn't linked to the list. Then I'll close this. Thanks!