mhamilton723 / STEGO

Unsupervised Semantic Segmentation by Distilling Feature Correspondences
MIT License
711 stars 142 forks source link

Model checkpointing only saving epoch 0 #66

Closed Holmes2002 closed 1 year ago

Holmes2002 commented 1 year ago

Why I run model I just receive epoch 0 ?? Where the orthers

mhamilton723 commented 1 year ago

I have the following checkpointing logic set up::

 ModelCheckpoint(
                dirpath=join(checkpoint_dir, name),
                every_n_train_steps=400,
                save_top_k=2,
                monitor="test/cluster/mIoU",
                mode="max",
            )

Perhaps your dataset is so large you didnt leave epoch 0 yet.

Holmes2002 commented 1 year ago

I solved it. Tks for your cmt !

FJGEODEV commented 7 months ago

I solved it. Tks for your cmt !

Could you please describe how you solved it? I got similar issue. Thanks.