LifeIsStrange

Results 246 comments of LifeIsStrange

Question on GPU Requirement when training locally

@KhalilMrini

Question on GPU Requirement when training locally

@matyascorvinus Noob question: what cloud do you recommend for training large neural networks?

requirements.txt is broken

no build error but crash on run python demo.py final Traceback (most recent call last): File "demo.py", line 7, in import coref_model as cm File "/home/stephane/coref2/3/e2e-coref/coref_model.py", line 17, in import...

requirements.txt is broken

todo try 1.12

Leverage deepspeed for much faster inference time

note Deepspeed Mii would be better https://github.com/microsoft/DeepSpeed-MII

Leverage deepspeed for much faster inference time

see the benchmark for stablediffusion (V1?): https://github.com/microsoft/DeepSpeed-MII/tree/main/examples/benchmark/txt2img

Improve your state of the art by using best activation function and best meta optimizer

@minimaxir it will not *necessarily* bring gains but it is still a low hanging fruit that should be tried.

Improve your state of the art by using best activation function and best meta optimizer

@digantamisra98 RAdam (not the full Ranger package) does not increase computational cost. I've read somewhere that Mish can be as efficient as Relu Maybe with https://github.com/thomasbrandon/mish-cuda?

Trying to make it work for user inputed sentences

@edobobo Yes I have downloaded a dataset and looked at the format. It is unclear to me how to transform a given sentence in the expected format... when to set...

Trying to make it work for user inputed sentences

Thanks a lot, is there a downside to requiring every words to be instance ?