Caleb Robinson
Caleb Robinson
fiona.transform.transform exhibits different behavior based on the underlying gdal library installed. ## Expected behavior and actual behavior. If I run `fiona.transform.transform("epsg:4326", "epsg:3857", [-75.71], [38.06])` I expect to get: `([-8427998.647958742], [4587905.271362515])`...
`show_anns(anns)` in [automatic_mask_generator_example.ipynb](https://github.com/facebookresearch/segment-anything/blob/main/notebooks/automatic_mask_generator_example.ipynb) calls `ax.imshow` for each mask in the output which can be very slow for large images. This PR changes `show_anns` to compute a single overlayed image that...
### Summary Currently the datamodules divide by 255 by default. This can confuse users who would expect no augmentations by default. We should make this WARNING level clear or not...
When we want to add a PASTIS datamodule we can start with this.
### Summary This issue is to track progress on implementing new pretrained weights from related literature into torchgeo: - [ ] Clay: [GitHub](https://github.com/Clay-foundation/model), [weights](https://huggingface.co/made-with-clay/Clay) @hfangcat - [x] CROMA: [GitHub](https://github.com/antofuller/CROMA) @nilsleh,...
TileDatasets
- Introduces a new class of datasets called TileDatasets that are indexed by filename, xoffset, yoffset, and patch size. - Implements samplers for these - Implements a L7IrishDataModule using this...
### Description I have a Sentinel 2 scene with the following files (e.g. in `./test_scene/`): ``` T36KVU_20210513T073609_B01_60m.tif T36KVU_20210513T073609_B02_10m.tif T36KVU_20210513T073609_B03_10m.tif T36KVU_20210513T073609_B04_10m.tif T36KVU_20210513T073609_B05_20m.tif T36KVU_20210513T073609_B06_20m.tif T36KVU_20210513T073609_B07_20m.tif T36KVU_20210513T073609_B08_10m.tif T36KVU_20210513T073609_B09_60m.tif T36KVU_20210513T073609_B11_20m.tif T36KVU_20210513T073609_B12_20m.tif T36KVU_20210513T073609_B8A_20m.tif T36KVU_20210513T073609_TCI_10m.tif ```...
The RwandaFieldBoundary dataset plots a mask even though this is unnecessary -- change the plot function to ignore the mask key in a sample if it doesn't exist. Also, the...
### Issue We create an IntersectionDataset like this: ``` train_image_ds = RasterDataset( 'data/processed/images/', ) train_mask_ds = RasterDataset( 'data/processed/masks/', ) train_mask_ds.is_image = False train_ds = train_image_ds & train_mask_ds ``` Here both...