Closed zimingzhong closed 2 years ago
Hi @zimingzhong ,
Thanks for your question, you should use the correct backbone for each model, just fixed the typo, should be
python -u test_lseg_zs.py --backbone clip_vitl16_384 --module clipseg_DPT_test_v2 --dataset fss \
--widehead --no-scaleinv --arch_option 0 --ignore_index 255 --fold 0 --nshot 0 \
--weights checkpoints/fss_l16.ckpt
Hope this helps!
I want to run
load it to <class 'modules.lseg_module_zs.LSegModuleZS'> and get the error.
size mismatch for net.scratch.layer4_rn.weight: copying a param with shape torch.Size([256, 1024, 3, 3]) from checkpoint, the shape in current model is torch.Size([256, 2048, 3, 3]).