Closed PatBall1 closed 1 year ago
Models do not generalise well across image scales. Can this be fixed by adjusting the data augmentation parameters? https://github.com/PatBall1/detectree2/blob/be81a1854959bd400bfd66c7d4f169f9aaafba48/detectree2/models/train.py#L155-L180
I suspect T.Resize((800, 800)) is not helping
T.Resize((800, 800))
See https://detectron2.readthedocs.io/en/latest/modules/data_transforms.html#detectron2.data.transforms.ResizeScale
Models do not generalise well across image scales. Can this be fixed by adjusting the data augmentation parameters? https://github.com/PatBall1/detectree2/blob/be81a1854959bd400bfd66c7d4f169f9aaafba48/detectree2/models/train.py#L155-L180