Use of different optimizers in SegFormer

NVlabs / SegFormer

Official PyTorch implementation of SegFormer

https://arxiv.org/abs/2105.15203

Other

2.58k stars 357 forks source link

Use of different optimizers in SegFormer #142

Open anirbala98 opened 11 months ago

anirbala98 commented 11 months ago

Keeping other parameters unchanged, if I try to use SGD instead of AdamW optimizer for Segformer model(modified with a Swin backbone), the results are so bad. I do get decent results with AdamW though. Only AdamW works with Segformer?