M3ViT
M3ViT copied to clipboard
VITA-Group
[NeurIPS 2022] “M³ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-design”, Hanxue Liang*, Zhiwen Fan*, Rishov Sarkar, Ziyu Jiang, Tianlong Chen,...
Hello, Thank you for your open source information. Will the hardware design also be available for reference? Or can you share some open-source baseline hardware design? Thanks in advance
I use this command to run the code `python main.py --config_env configs/env.yml --config_exp configs/$DATASET/$MODEL.yml` since I can only install fastmoe==0.1.2. However, when I run with the command above. the following...
hi , i am so sorry to ask you a last checkpoint and its pretrain file ,for vitmoe ,because i can not train my model good like yours, very very...
Metadata
Owner
Metadata
[NeurIPS 2022] “M³ViT: Mixture-of-Experts Vision Transformer for Efficient Multi-task Learning with Model-Accelerator Co-design”, Hanxue Liang*, Zhiwen Fan*, Rishov Sarkar, Ziyu Jiang, Tianlong Chen,...