google-research / mint

Multi-modal Content Creation Model Training Infrastructure including the FACT model (AI Choreographer) implementation.
Apache License 2.0
497 stars 85 forks source link

CUDA, Tensorflow, nvcc versions? #34

Open jgallegov opened 2 years ago

jgallegov commented 2 years ago

Hello,

which CUDA, tensorflow, nvcc versions are you using? I am having problems with tensorflow. It does not recognise my GPUs and it should be related to version compatibility.

YiAthena commented 2 years ago

tensorflow 2.4, cuda 11.0, cudnn 8.0 work for me

Concernin9 commented 2 years ago

tensorflow 2.4, cuda 11.0, cudnn 8.0 work for me

@YiAthena Hello, the result of the code is the data in NPY format. I found that to form an smpl model, do I need to convert NPY format into PKL format first? I want to visualize the results. Is this the process? Looking forward to your reply, thank you very much!

YiAthena commented 2 years ago

do I need to convert NPY format into PKL format first? -> in my case, no. It depends on your reconstruction method.

jgallegov commented 2 years ago

Hello, then, what how is the .npy content generated? In .pkl is very clear, but this npy does not have any fields.