something wrong with vit-lite

SHI-Labs / Compact-Transformers

Escaping the Big Data Paradigm with Compact Transformers, 2021 (Train your Vision Transformers in 30 mins on CIFAR-10 with a single GPU!)

https://arxiv.org/abs/2104.05704

Apache License 2.0

495 stars 77 forks source link

something wrong with vit-lite #70

Closed EatonL closed 1 year ago

EatonL commented 1 year ago

When I run the model named "vit_7_4_32" , this error will occur: TypeError: type object got multiple values for keyword argument 'positional_embedding'

This is the command I run： CUDA_VISIBLE_DEVICES=1 python train.py -c configs/datasets/cifar100.yml --model vit_7_4_32 myPicfolder/cifar-100 --output myResultfolder/ViTlite --experiment vit_7_4_32_2022_cifar100 --log-wandb --dataset cifar100

Looking forward to your reply，thks

EatonL commented 1 year ago

I already know how to solve this bug

alihassanijr commented 1 year ago

Thank you for bringing this to our attention. PR #71 will resolve this issue.