search

zhengchen1999 / DAT

PyTorch code for our ICCV 2023 paper "Dual Aggregation Transformer for Image Super-Resolution"
Apache License 2.0
350 stars 27 forks source link

Training Process and GPU #13

Open Zhihao-chen-666 opened 9 months ago

Zhihao-chen-666 commented 9 months ago

Hello, I use 4 3090 gpu to train the model, but it throw error "cuda memory is out of range". Could you please tell me what's the type of gpu you use? and how long dose the training process will take? Thank you!

zhengchen1999 commented 8 months ago

Hi. Sorry for being late. For DAT, 4 GPUs with a memory $\geq$ 40 Gb for training. I train the model on 4 A100s, and it takes 3 days.