Training with model (KITTI Swin-Large) ，why the backbone uses swin_base_v2 ,why not swin_large_v2 ?

Question

Training with model (KITTI Swin-Large) ，why the backbone uses swin_base_v2 ,why not swin_large_v2 ?

Closed this issue a year ago · 2 comments

When evaluate with model with Swin-Base and Swin-Large , i found that the command uses same ckpt_dir " --ckpt_dir ckpt/kitti_swin_base.ckpt" and the "--backbone" still swin_base_v2. It's reasonable？

Answer 1 · 2023-03-07T03:41:25.000Z

use this instead

python test.py --dataset kitti --kitti_crop garg_crop --data_path ../data/ --max_depth 80.0 --max_depth_eval 80.0 --backbone swin_large_v2 --depths 2 2 18 2 --num_filters 32 32 32 --deconv_kernels 2 2 2 --window_size 22 22 22 11 --pretrain_window_size 12 12 12 6 --use_shift True True False False --flip_test --shift_window_test --shift_size 16 --do_evaluate --ckpt_dir ckpt/kitti_swin_large.ckpt

Answer 2 · 2023-03-27T05:05:23.000Z

Sorry for the typo. It has been fixed.