search

jzhangbs / Vis-MVSNet

Visibility-aware Multi-view Stereo Network
MIT License
235 stars 27 forks source link

Fail to extract the pretrained model 20000.tar #4

Closed mazeyu closed 3 years ago

mazeyu commented 3 years ago

Why do I fail to extract the pretrained model 20000.tar? Thanks

XYZ-qiyh commented 3 years ago

I met the same problem. Could you please upload the uncompressed model weights? @jzhangbs image

jzhangbs commented 3 years ago

You don't need to extract it. It's my mistake to consider that pytorch will save the model as a tarball. You can use the pretrained model by --load_path ./pretrained_model/vis

mazeyu commented 3 years ago

Thanks. It works.