OWOD/detectron2/data/datasets
Joseph 81516e5543 - Saving feature store
- Implemented Absolute OSE error and initial WI
2020-11-09 10:34:45 +05:30
..
README.md Detectron2 2020-09-29 05:47:18 +05:30
__init__.py Detectron2 2020-09-29 05:47:18 +05:30
builtin.py - Saving feature store 2020-11-09 10:34:45 +05:30
builtin_meta.py Detectron2 2020-09-29 05:47:18 +05:30
cityscapes.py Detectron2 2020-09-29 05:47:18 +05:30
cityscapes_panoptic.py Detectron2 2020-09-29 05:47:18 +05:30
coco.py Detectron2 2020-09-29 05:47:18 +05:30
lvis.py Detectron2 2020-09-29 05:47:18 +05:30
lvis_v0_5_categories.py Detectron2 2020-09-29 05:47:18 +05:30
lvis_v1_categories.py Detectron2 2020-09-29 05:47:18 +05:30
pascal_voc.py Updated the evaluation display. Ready for t1_std_frcnn. 2020-11-04 10:35:46 +05:30
pascal_voc_basic.py - Adding new imagesets 2020-11-04 09:34:21 +05:30
register_coco.py Detectron2 2020-09-29 05:47:18 +05:30
voc_style_coco.py - Changes to data-modelling 2020-11-03 16:59:41 +05:30

README.md

Common Datasets

The dataset implemented here do not need to load the data into the final format. It should provide the minimal data structure needed to use the dataset, so it can be very efficient.

For example, for an image dataset, just provide the file names and labels, but don't read the images. Let the downstream decide how to read.