FreemanG comments

Results 7 comments of


                                            FreemanG

How to ensure reproducibly data augmentation.

I did experiments on the basis of the code - IXI_HH_sex_classification_resnet. I found that setting random seed in the **train** function did not ensure reproducible data augmentation: the results of...

How to ensure reproducibly data augmentation.

I might by far be able to solve this problem by setting np.random.seed within the **reader_fn**. Specifically, I passed a seed to **reader_fn** through **reader_params**, and recreate **train_input_fn** with a...

multi-gpu training

It seems it works with multi-GPUs when you put the "assign_adain_params" function into the "Decoder" class.

multi-gpu training

Great :+1:

Poses are camera-to-world?

So I have to change this [line](https://github.com/Kai-46/nerfplusplus/blob/ebf2f3e75fd6c5dfc8c9d0b533800daaf17bd95f/colmap_runner/normalize_cam_dict.py#L55) > return np.linalg.inv(C2W) to > return C2W Is this right? The experiments seem better so far.

Question about "Motion-X format to HumanML3D format"

Hi @JunyuShi02. Have you solved this problem? I stuck here too.

Question about "Motion-X format to HumanML3D format"

> > Hi [@JunyuShi02](https://github.com/JunyuShi02). Have you solved this problem? I stuck here too. > > You can try downloading the new version provided by the author on the Huggingface. Thanks...