Shuhao Cao

https://scaomath.github.io

Overland Park An amateur computational PDE guy recently jumping on the bandwagon of deep learning. Andromeda (the galaxy, my avatar) is a cat living with our family.

Results 17 comments of


                                            Shuhao Cao

Loss function

The simplest way is to write your own: ```python def loss1(pred, target): '''first loss here''' return ... def loss2(pred, target): '''second loss here''' return ... def loss(pred, target): return a*loss1(pred,...

AttributeError: 'tuple' object has no attribute 'size'

https://github.com/TylerYep/torchinfo/blob/ed0a7aa3f61171d2a7f4af8e7958124304c7ef91/torchinfo/layer_info.py#L127-L134 @TylerYep What is the current dev roadmap for this function `calculate_size`? Would you take a PR that adds this functionality supporting dicts containing lists or tuples?

One complex parameter should count as two params

> Thanks for looking into this. Could you provide a small example model + output to illustrate your point? That will help validate this fix works I added a test....

Why it only works for matrices whose row and column are smaller than 32

https://github.com/KinglittleQ/torch-batch-svd/blob/c0a96119187f7d55f939d2ff2b92942c6d6ca930/torch_batch_svd/batch_svd.py#L19-L21 Just change this assertion and you are good to go (of course it depends on single/double precision and the VRAM on your GPU).

How to fix errors in code ex1_burgers_eval.ipynb: NameError: name 'FourierTransformer' is not defined

Thanks. Let me clean some NameError in evaluation.

Fixed Taylor Green decay coefficient and added an example

@shoyer

Are all the data DNS(direct numerical simulation)?

I wondered this question as well, yet it seems that no one answers questions from months ago #85 The paper did not specify what is the implementation for the interface...

‹
1
2