Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Figure plotting with Eurosat & classification trainer gives blank images #2139

Closed
robmarkcole opened this issue Jun 27, 2024 · 3 comments
Closed
Labels
datasets Geospatial or benchmark datasets

Comments

@robmarkcole
Copy link
Contributor

Description

As title:

image

Steps to reproduce

datamodule = EuroSATDataModule(batch_size=64, num_workers=8, bands=EuroSAT.BAND_SETS["all"])
datamodule.setup(stage="fit")

task = ClassificationTask(
    model="resnet18",
    in_channels=13,
    num_classes=10,
    loss="ce",
    lr=0.001,
)

logger = TensorBoardLogger("tb_logs", name="my_eurosat")

trainer = L.Trainer(
    max_epochs=10,
    logger=logger
)
trainer.fit(task, datamodule)

Version

master

@robmarkcole
Copy link
Contributor Author

Appears the issue is not with the basic plotting:

image

@adamjstewart
Copy link
Collaborator

I'm guessing this is related to #1263? That is, dataset plotting works, but data module plotting (in which data augmentation has been applied) does not? The solution for this dataset is the same as for all other datasets where this issue occurs, and is described in #1263.

@adamjstewart adamjstewart added the datasets Geospatial or benchmark datasets label Jun 29, 2024
@robmarkcole
Copy link
Contributor Author

Noted this is likely effecting many datasets and due to the norm. Closing and continuing on the linked issue

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
datasets Geospatial or benchmark datasets
Projects
None yet
Development

No branches or pull requests

2 participants