v0.33.0

What's Changed

• Add DINOv3 finetuning for image segmentation by @jayakumarpujar in #589
• Enhance geographic coordinate conversion in object detection by @sobakrim in #601
• Fix typos, missing spaces, and outdated instructions in docs by @jayakumarpujar in #602
• Add train_transforms and val_transforms to train_timm_segmentation_model by @giswqs in #606
• Fix pandas unavailable error in QGIS DeepForest subprocess mode by @giswqs in #607
• Improve tile overview image readability and add mask background by @giswqs in #614
• Fix training tiles display and notebook path error by @giswqs in #615
• [pre-commit.ci] pre-commit autoupdate by @pre-commit-ci[bot] in #613
• Bump docker/build-push-action from 6 to 7 by @dependabot[bot] in #611
• Bump docker/login-action from 3 to 4 by @dependabot[bot] in #610
• Bump docker/metadata-action from 5 to 6 by @dependabot[bot] in #609
• fix: correct inference_on_geotiff return type to Tuple[str, float] by @sobakrim in #612
• Add memory-efficient tiled inference with spline blending and D4 TTA by @jayakumarpujar in #603
• Add image_subdir and mask_subdir params to display_training_tiles by @giswqs in #617
• Handle list and .tiff inputs in object_detection_batch by @sobakrim in #616
• Fix .tiff output filenames and update docstring in object_detection_batch by @giswqs in #618

New Contributors

• @jayakumarpujar made their first contribution in #580
• @sobakrim made their first contribution in #588

Full Changelog: v0.31.0...v0.33.0