Search results

训练mscoco时的’val‘文件夹

Status: Open.

#30 In baofff/U-ViT;

· Maxsim111 opened on May 7, 2025

Why different fid_stats are available for ImageNet dataset for different resolutions? How to generate fid_stats for Celeba dataset for resolutions other than 64x64?

Status: Open.

#29 In baofff/U-ViT;

· LakshmikarPolamreddy opened on Nov 14, 2024

training gets slower and stopped

Status: Open.

#28 In baofff/U-ViT;

· forever208 opened on Aug 2, 2024

Some NCCL operations have failed or timed out

Status: Open.

#27 In baofff/U-ViT;

· forever208 opened on Aug 2, 2024

Visualing attention layers (q-k)

Status: Open.

#26 In baofff/U-ViT;

· unlugi opened on Jul 18, 2024

ImportError: cannot import name 'MSCOCODatabase' from 'datasets'

Status: Open.

#25 In baofff/U-ViT;

· lihao-doc opened on Jul 11, 2024

Autoencoder

Status: Open.

#24 In baofff/U-ViT;

· Rowerliu opened on May 9, 2024

无法训练

Status: Open.

#18 In baofff/U-ViT;

· xiaoxiaodadada opened on Oct 29, 2023

why the scale_factor of autoencoder used in MSCOCO configs is 0.2301?

Status: Open.

#17 In baofff/U-ViT;

· moonbaby9 opened on Jul 25, 2023

multiple nodes distributed training

Status: Open.

#16 In baofff/U-ViT;

· ilovecv opened on Jun 7, 2023

the difference between train_ldm_discrete.py with train_t2i_discrete.py

Status: Open.

#15 In baofff/U-ViT;

· moonbaby9 opened on Jun 5, 2023

训练时accelerate报错

Status: Open.

#12 In baofff/U-ViT;

· SYZhang0805 opened on May 2, 2023