BigEarthNet Trainers #211
BigEarthNet Trainers #211
Conversation
isaaccorley
force-pushed
the
trainers/bigearthnet
branch
from
c4d243e to
d6e548b
Compare
|
I refactored test_tasks to use a dummy dataset and datamodule so we can test for varying number of channels but avoid having to repeat tests for bigearthnet and so2sat datamodule specific args. Let me know if this works. |
|
You are right thought that there is not a good case for making a separate task for each classification dataset. We can definitely remove them if that's the route you want to take. |
| DataLoader.__module__ = "torch.utils.data" | ||
|
|
||
|
|
||
| class BigEarthNetClassificationTask(MultiLabelClassificationTask): |
There was a problem hiding this comment.
I think getting rid of this class is the last remaining thing to do before this can be merged.
There was a problem hiding this comment.
Personally I would refactor ClassificationTask and MultiLabelClassificationTask first in separate PRs to keep this PR from getting any bigger.
There was a problem hiding this comment.
We should merge this because I still need to refactor the num classes as an input arg before I can remove it. If I do that then I will need to remove/refactor all other classification tasks as well.
* add additional bigearthnet test data for train/val/test split * update bigearthnet dataset length test * add MultiLabelClassificationTask * add BigEarthNet trainer and datamodule * add bigearthnet and multilabelclassificationtask tests * mypy and format * add estimated band min/max values for normalization * softmax outputs to correctly compute metrics * update min/max stats for 100k samples * organize imports in torchgeo.trainers.__init__.py * clean up fixtures in test_tasks.py * added bigearthnet to train.py * format * move fixtures into class methods * consolidate bigearthnet fixtures * refactor tasks tests * add scope=class * style/mypy fixes * mypy fixes
torchgeo.trainers.BigEarthNetClassificationTaskandtorchgeo.trainers.BigEarthNetDataModuletorchgeo.trainers.tasks.MultiLabelClassficationTask. This differs fromClassificationTaskby usingtorch.nn.BCEWithLogitsLossand modifies the metrics to handle multilabel outputsTODO: