[Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images #3695

ZwwWayne · 2020-09-05T06:24:37Z

This PR fixes the issue of training with empty-GT images when specifying classes, continuing on #3133.

In the original implementation, get_subset_by_classes is not related to the flag self.filter_empty_gt. This brings ambiguous behavior and potential bugs when specifying dataset classes. More specifically, CocoDataset filters empty gt images in two places, one is in _filter_img, the other one is in get_subset_by_classes. Other datasets only filter empty gt images in get_subset_by_classes, but this will be triggered by specifying classes rather than self.filter_empty_gt=True.

The goal of this PR is to make the behavior of filter empty GT independent to specifying classes. After this PR, if filter_empty_gt=False, no matter whether the classes are specified in a dataset, the dataset will use all the images in the annotations. If filter_empty_gt=True, no matter whether the classes are specified, the dataset will check the images and filter out images containing no GT boxes.

This PR also changes the logic during test_mode. Previously, the dataset will filter out empty-GT images no matter whether it is in test mode. After this PR, it only filters out images in training, i.e., if test_mode=True, it will test all the images given in the annotations. This behavior is more commonly need as mentioned in #3545, and is more consistent with before that the dataset will not filter out small images in test mode. Therefore, the users should be responsible for the data filtering/cleaning process for the test dataset.

Fix #3023, Fix #2965, Fix #3545

…detection into fix_coco_loader_empty_images Conflicts: .gitignore tests/test_dataset.py

codecov · 2020-09-05T09:20:49Z

Codecov Report

Merging #3695 into master will decrease coverage by 0.06%.
The diff coverage is 29.41%.

@@            Coverage Diff             @@
##           master    #3695      +/-   ##
==========================================
- Coverage   62.48%   62.41%   -0.07%     
==========================================
  Files         219      219              
  Lines       16256    16257       +1     
  Branches     2753     2753              
==========================================
- Hits        10157    10147      -10     
- Misses       5603     5616      +13     
+ Partials      496      494       -2

Flag	Coverage Δ
#unittests	`62.41% <29.41%> (-0.07%)`	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
mmdet/datasets/cityscapes.py	`12.25% <0.00%> (-0.58%)`	⬇️
mmdet/datasets/custom.py	`48.87% <ø> (-2.56%)`	⬇️
mmdet/datasets/xml_style.py	`16.50% <5.88%> (-2.69%)`	⬇️
mmdet/datasets/coco.py	`45.76% <100.00%> (+0.12%)`	⬆️

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update d460a53...96715cd. Read the comment docs.

…oco_loader_empty_images

mmdet/datasets/coco.py

mmdet/datasets/custom.py

mmdet/datasets/coco.py

mmdet/datasets/custom.py

…oco_loader_empty_images

hellock · 2020-09-30T02:28:51Z

Task linked: CU-4hu3hu PR #3695

* Add doc of modify loss (#3777) * add doc of modify loss * minor fix * change name * improve the function of simple_test_bboxes (#3853) * update * fix lint error * add bbox_pred_split in sabl_head * fix lint error * fix yapf * fix build error * fix yapf * Clean background_labels in the dense heads (#3221) * Clean background_labels in the dense heads * clean format * Reformat * Fix CI bug * fix fsaf and the comments * fix newly added heads * fix yolact * complete comments * fix rpn transforming bug in two stage networks (#3754) * fix rpn transforming bug in two_stage * [Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images (#3695) * Add regression test and test data * Add fix for empty gt images * Trigger CI build * refactor get_subset_by_classes * reformat and fix docstring * [refactor]: move get_subset_by_classes to _filter_imgs * make img_ids consistent * resolve comments * simplify logics * add warning * add warning * add warning Co-authored-by: mmeendez8 <miguelmndez@gmail.com> * Fix nonzero in NMS for PyTorch 1.6.0 (#3867) * Support TTA of ATSS, FCOS, YOLOv3 (#3844) * Support TTA of ATSS, FCOS, YOLOv3 * Add comment * [Docs] Fix typo in docs/tutorials/new_dataset.md (#3876) * [Docs] Remove duplicate content in docs/config.md (#3875) * [Enhance]: Convert mask to bool before using it as img's index for robustness and speedup (#3870) * Convert mask to bool before using it as img's index * lint * Rewrite code in more concise way * Fix typo in bbox_flip (#3886) * fix the API change bug of PAA (#3883) * fix cv2 import error of ligGL.so.1 (#3891) * [enhance]: Improve documentation of modules and dataset customization (#3821) * update tutorials for changing training settings * update * Update details of hook * resolve comments * reformat * Update module documentations * add notes for classes specification * support to use pytorch 1.6 in docker (#3905) * support to use pytorch 1.6 in docker * update mmcv requirement * update docs * Add missing notes in data customization (#3906) * [Fix]: fix mask rcnn training stuck problem when there is no positive rois (#3713) * Fix mask rcnn stuck problem when there is no positive rois * support non pos inference in cascade methods, link CU-49tawu * print mmcv version in CI * use mmcv repo to check wrappers * change cpu build * upgrade mmcv requirements and change ci back * use pre-built whl in CI * Bump to v2.5.0 (#3879) * bump to v2.5.0 * update PR * resolve comments * clean unnecessary modification * update bc breaking * rephrase bc-breaking * rephrase * Added `generate_inputs_and_wrap_model` function for pytorch2onnx (#3857) * added `generate_inputs_and_wrap_model` function for pytorch2onnx * undo debugging changes * added model.cpu().eval() * refactored onnx * fix verify * move `generate_inputs_and_wrap_model` into mmdet.core.export * use `generate_inputs_and_wrap_model` in tools/pytorch2onnx.py * typo (#3917) * Edit mmdet.core.export docstring (#3912) * edit mmdet.core.export docstring * update doctest * fix small bugs * supports for HungarianMatchAssigner, add bbox_cxcywh_to_xyxy and bbox_xyxy_to_cxcywh for bbox transform * format box-wise related giou calculating as a function and implement it in iou2d_calculater.py * supports for BboxGIoU2D and re-implements giou_loss using bbox_gious * remove unnecessary * reformat * reformat docstring * reformat * rename * supports for giou calculating in BboxOverlaps2D, and re-implements giou_loss using bbox_overlaps * fix sabl validating bug in cascade_rcnn (#3913) * reformat * move giou related unit test from test_losses.py to test_iou2d_calculator.py * reformat * Avoid division by zero in PAA head when num_pos=0 * [Fix]: Avoid division by zero in PAA head when num_pos=0 (#3938) * explicitly add mode in giou_loss * Add supports for giou calculation in BboxOverlaps2D, and re-implements giou_loss using bbox_overlaps (#3936) * supports for BboxGIoU2D and re-implements giou_loss using bbox_gious * remove unnecessary * reformat * reformat docstring * reformat * rename * supports for giou calculating in BboxOverlaps2D, and re-implements giou_loss using bbox_overlaps * reformat * move giou related unit test from test_losses.py to test_iou2d_calculator.py * reformat * Avoid division by zero in PAA head when num_pos=0 * explicitly add mode in giou_loss * Add supports for giou calculation in BboxOverlaps2D, and add iou_calculator in hungarian_assigner * rename hungarian_match_assigner as hungarian_assigner * fix init * reformat docstring * Avoid division by zero in PAA head when num_pos=0 * fix cpu (#3948) * add mode for iou_calculator and make giou cost as a default case * make mode as a param in iou_calculator * reformat docsting * make iou_mode outside of iou_calculator Co-authored-by: Cao Yuhang <yhcao6@gmail.com> Co-authored-by: yuzhj <31239926+yuzhj@users.noreply.github.com> Co-authored-by: Wenwei Zhang <40779233+ZwwWayne@users.noreply.github.com> Co-authored-by: mmeendez8 <miguelmndez@gmail.com> Co-authored-by: Yosuke Shinya <42844407+shinya7y@users.noreply.github.com> Co-authored-by: LabMemNo003 <LabMemNo003@gmail.com> Co-authored-by: Ye Liu <yeliudev@outlook.com> Co-authored-by: aboettcher <aboettcher@users.noreply.github.com> Co-authored-by: Ryan Li <xlide@connect.ust.hk> Co-authored-by: Jiangmiao Pang <pangjiangmiao@gmail.com>

* supports for DETR transformer * unit test with small data to avoid out of memory in CI * use batch size 1 for unit test to avoid out of memory * move transformer into utils folder and use more small data for unit test * reformat docstring * add more detailed docstring * reforamt * reformat and add build_transformer (#3866) * Supports for DETR position embedding (#3850) * supports for DETR position embedding * reformat docstring * reformat * move transformer_head folder into utils * remove unnecessary comments * reformat * reformat * reformat docstring * reformat * add optional for default parameters * add comments for unit test * avoid inplace in MultiheadAttention and add optional add_residual for FFN * Supports for DETR inference (#3941) * fix forward in FFN * supports detr inference mode with batchsize 1 * add docstring for DETR in detr.py * reformat and add doctest for TransformerHead * reformat, and move the mask construction from simple_test in detr.py to the forward process of transformer_head * rename detr_r50_1x_coco as detr_r50_8x4_1x_coco in config * reformat * mask input_img_shape in img_meta as a general modification of base_detector forward_test * reformat * Supports for DETR hungarian matcher. (#3929) * Add doc of modify loss (#3777) * add doc of modify loss * minor fix * change name * improve the function of simple_test_bboxes (#3853) * update * fix lint error * add bbox_pred_split in sabl_head * fix lint error * fix yapf * fix build error * fix yapf * Clean background_labels in the dense heads (#3221) * Clean background_labels in the dense heads * clean format * Reformat * Fix CI bug * fix fsaf and the comments * fix newly added heads * fix yolact * complete comments * fix rpn transforming bug in two stage networks (#3754) * fix rpn transforming bug in two_stage * [Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images (#3695) * Add regression test and test data * Add fix for empty gt images * Trigger CI build * refactor get_subset_by_classes * reformat and fix docstring * [refactor]: move get_subset_by_classes to _filter_imgs * make img_ids consistent * resolve comments * simplify logics * add warning * add warning * add warning Co-authored-by: mmeendez8 <miguelmndez@gmail.com> * Fix nonzero in NMS for PyTorch 1.6.0 (#3867) * Support TTA of ATSS, FCOS, YOLOv3 (#3844) * Support TTA of ATSS, FCOS, YOLOv3 * Add comment * [Docs] Fix typo in docs/tutorials/new_dataset.md (#3876) * [Docs] Remove duplicate content in docs/config.md (#3875) * [Enhance]: Convert mask to bool before using it as img's index for robustness and speedup (#3870) * Convert mask to bool before using it as img's index * lint * Rewrite code in more concise way * Fix typo in bbox_flip (#3886) * fix the API change bug of PAA (#3883) * fix cv2 import error of ligGL.so.1 (#3891) * [enhance]: Improve documentation of modules and dataset customization (#3821) * update tutorials for changing training settings * update * Update details of hook * resolve comments * reformat * Update module documentations * add notes for classes specification * support to use pytorch 1.6 in docker (#3905) * support to use pytorch 1.6 in docker * update mmcv requirement * update docs * Add missing notes in data customization (#3906) * [Fix]: fix mask rcnn training stuck problem when there is no positive rois (#3713) * Fix mask rcnn stuck problem when there is no positive rois * support non pos inference in cascade methods, link CU-49tawu * print mmcv version in CI * use mmcv repo to check wrappers * change cpu build * upgrade mmcv requirements and change ci back * use pre-built whl in CI * Bump to v2.5.0 (#3879) * bump to v2.5.0 * update PR * resolve comments * clean unnecessary modification * update bc breaking * rephrase bc-breaking * rephrase * Added `generate_inputs_and_wrap_model` function for pytorch2onnx (#3857) * added `generate_inputs_and_wrap_model` function for pytorch2onnx * undo debugging changes * added model.cpu().eval() * refactored onnx * fix verify * move `generate_inputs_and_wrap_model` into mmdet.core.export * use `generate_inputs_and_wrap_model` in tools/pytorch2onnx.py * typo (#3917) * Edit mmdet.core.export docstring (#3912) * edit mmdet.core.export docstring * update doctest * fix small bugs * supports for HungarianMatchAssigner, add bbox_cxcywh_to_xyxy and bbox_xyxy_to_cxcywh for bbox transform * format box-wise related giou calculating as a function and implement it in iou2d_calculater.py * supports for BboxGIoU2D and re-implements giou_loss using bbox_gious * remove unnecessary * reformat * reformat docstring * reformat * rename * supports for giou calculating in BboxOverlaps2D, and re-implements giou_loss using bbox_overlaps * fix sabl validating bug in cascade_rcnn (#3913) * reformat * move giou related unit test from test_losses.py to test_iou2d_calculator.py * reformat * Avoid division by zero in PAA head when num_pos=0 * [Fix]: Avoid division by zero in PAA head when num_pos=0 (#3938) * explicitly add mode in giou_loss * Add supports for giou calculation in BboxOverlaps2D, and re-implements giou_loss using bbox_overlaps (#3936) * supports for BboxGIoU2D and re-implements giou_loss using bbox_gious * remove unnecessary * reformat * reformat docstring * reformat * rename * supports for giou calculating in BboxOverlaps2D, and re-implements giou_loss using bbox_overlaps * reformat * move giou related unit test from test_losses.py to test_iou2d_calculator.py * reformat * Avoid division by zero in PAA head when num_pos=0 * explicitly add mode in giou_loss * Add supports for giou calculation in BboxOverlaps2D, and add iou_calculator in hungarian_assigner * rename hungarian_match_assigner as hungarian_assigner * fix init * reformat docstring * Avoid division by zero in PAA head when num_pos=0 * fix cpu (#3948) * add mode for iou_calculator and make giou cost as a default case * make mode as a param in iou_calculator * reformat docsting * make iou_mode outside of iou_calculator Co-authored-by: Cao Yuhang <yhcao6@gmail.com> Co-authored-by: yuzhj <31239926+yuzhj@users.noreply.github.com> Co-authored-by: Wenwei Zhang <40779233+ZwwWayne@users.noreply.github.com> Co-authored-by: mmeendez8 <miguelmndez@gmail.com> Co-authored-by: Yosuke Shinya <42844407+shinya7y@users.noreply.github.com> Co-authored-by: LabMemNo003 <LabMemNo003@gmail.com> Co-authored-by: Ye Liu <yeliudev@outlook.com> Co-authored-by: aboettcher <aboettcher@users.noreply.github.com> Co-authored-by: Ryan Li <xlide@connect.ust.hk> Co-authored-by: Jiangmiao Pang <pangjiangmiao@gmail.com> * Re-implements RandomCrop to support different crop_type (#4093) * re-implements RandomCrop to support different crop_type * add assertion for relative related crop_type * reformat, and add image shape assertion for unit test * reformat * Supports for DETR training mode in process. (#3963) * add input_img_shape in img_meta, and supports training mode toy * toy commit * toy commit * re-organize and support for detr training mode * add unit test for detr training, testing, and transformer_head * add more unit test for transformer * reformat * remove unnecessary * use small transformer model for unit test * reformat docstring * reformat * reformat * Supports DETR e150 config (#4197) * rename e300 config as e150 config * reformat * add comments on override option in Resize * add comments on override option in Resize * position embeddingto positional encoding * fix unit tests * fix registry name bug * rename file Co-authored-by: v-qjqs <qjliqiaofei@gmail.com> Co-authored-by: Qiaofei Li <34116221+v-qjqs@users.noreply.github.com> Co-authored-by: Cao Yuhang <yhcao6@gmail.com> Co-authored-by: yuzhj <31239926+yuzhj@users.noreply.github.com> Co-authored-by: mmeendez8 <miguelmndez@gmail.com> Co-authored-by: Yosuke Shinya <42844407+shinya7y@users.noreply.github.com> Co-authored-by: LabMemNo003 <LabMemNo003@gmail.com> Co-authored-by: Ye Liu <yeliudev@outlook.com> Co-authored-by: aboettcher <aboettcher@users.noreply.github.com> Co-authored-by: Ryan Li <xlide@connect.ust.hk> Co-authored-by: Jiangmiao Pang <pangjiangmiao@gmail.com>

* supports for DETR transformer * unit test with small data to avoid out of memory in CI * use batch size 1 for unit test to avoid out of memory * move transformer into utils folder and use more small data for unit test * reformat docstring * add more detailed docstring * reforamt * reformat and add build_transformer (#3866) * Supports for DETR position embedding (#3850) * supports for DETR position embedding * reformat docstring * reformat * move transformer_head folder into utils * remove unnecessary comments * reformat * reformat * reformat docstring * reformat * add optional for default parameters * add comments for unit test * avoid inplace in MultiheadAttention and add optional add_residual for FFN * Supports for DETR inference (#3941) * fix forward in FFN * supports detr inference mode with batchsize 1 * add docstring for DETR in detr.py * reformat and add doctest for TransformerHead * reformat, and move the mask construction from simple_test in detr.py to the forward process of transformer_head * rename detr_r50_1x_coco as detr_r50_8x4_1x_coco in config * reformat * mask input_img_shape in img_meta as a general modification of base_detector forward_test * reformat * Supports for DETR hungarian matcher. (#3929) * Add doc of modify loss (#3777) * add doc of modify loss * minor fix * change name * improve the function of simple_test_bboxes (#3853) * update * fix lint error * add bbox_pred_split in sabl_head * fix lint error * fix yapf * fix build error * fix yapf * Clean background_labels in the dense heads (#3221) * Clean background_labels in the dense heads * clean format * Reformat * Fix CI bug * fix fsaf and the comments * fix newly added heads * fix yolact * complete comments * fix rpn transforming bug in two stage networks (#3754) * fix rpn transforming bug in two_stage * [Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images (#3695) * Add regression test and test data * Add fix for empty gt images * Trigger CI build * refactor get_subset_by_classes * reformat and fix docstring * [refactor]: move get_subset_by_classes to _filter_imgs * make img_ids consistent * resolve comments * simplify logics * add warning * add warning * add warning Co-authored-by: mmeendez8 <miguelmndez@gmail.com> * Fix nonzero in NMS for PyTorch 1.6.0 (#3867) * Support TTA of ATSS, FCOS, YOLOv3 (#3844) * Support TTA of ATSS, FCOS, YOLOv3 * Add comment * [Docs] Fix typo in docs/tutorials/new_dataset.md (#3876) * [Docs] Remove duplicate content in docs/config.md (#3875) * [Enhance]: Convert mask to bool before using it as img's index for robustness and speedup (#3870) * Convert mask to bool before using it as img's index * lint * Rewrite code in more concise way * Fix typo in bbox_flip (#3886) * fix the API change bug of PAA (#3883) * fix cv2 import error of ligGL.so.1 (#3891) * [enhance]: Improve documentation of modules and dataset customization (#3821) * update tutorials for changing training settings * update * Update details of hook * resolve comments * reformat * Update module documentations * add notes for classes specification * support to use pytorch 1.6 in docker (#3905) * support to use pytorch 1.6 in docker * update mmcv requirement * update docs * Add missing notes in data customization (#3906) * [Fix]: fix mask rcnn training stuck problem when there is no positive rois (#3713) * Fix mask rcnn stuck problem when there is no positive rois * support non pos inference in cascade methods, link CU-49tawu * print mmcv version in CI * use mmcv repo to check wrappers * change cpu build * upgrade mmcv requirements and change ci back * use pre-built whl in CI * Bump to v2.5.0 (#3879) * bump to v2.5.0 * update PR * resolve comments * clean unnecessary modification * update bc breaking * rephrase bc-breaking * rephrase * Added `generate_inputs_and_wrap_model` function for pytorch2onnx (#3857) * added `generate_inputs_and_wrap_model` function for pytorch2onnx * undo debugging changes * added model.cpu().eval() * refactored onnx * fix verify * move `generate_inputs_and_wrap_model` into mmdet.core.export * use `generate_inputs_and_wrap_model` in tools/pytorch2onnx.py * typo (#3917) * Edit mmdet.core.export docstring (#3912) * edit mmdet.core.export docstring * update doctest * fix small bugs * supports for HungarianMatchAssigner, add bbox_cxcywh_to_xyxy and bbox_xyxy_to_cxcywh for bbox transform * format box-wise related giou calculating as a function and implement it in iou2d_calculater.py * supports for BboxGIoU2D and re-implements giou_loss using bbox_gious * remove unnecessary * reformat * reformat docstring * reformat * rename * supports for giou calculating in BboxOverlaps2D, and re-implements giou_loss using bbox_overlaps * fix sabl validating bug in cascade_rcnn (#3913) * reformat * move giou related unit test from test_losses.py to test_iou2d_calculator.py * reformat * Avoid division by zero in PAA head when num_pos=0 * [Fix]: Avoid division by zero in PAA head when num_pos=0 (#3938) * explicitly add mode in giou_loss * Add supports for giou calculation in BboxOverlaps2D, and re-implements giou_loss using bbox_overlaps (#3936) * supports for BboxGIoU2D and re-implements giou_loss using bbox_gious * remove unnecessary * reformat * reformat docstring * reformat * rename * supports for giou calculating in BboxOverlaps2D, and re-implements giou_loss using bbox_overlaps * reformat * move giou related unit test from test_losses.py to test_iou2d_calculator.py * reformat * Avoid division by zero in PAA head when num_pos=0 * explicitly add mode in giou_loss * Add supports for giou calculation in BboxOverlaps2D, and add iou_calculator in hungarian_assigner * rename hungarian_match_assigner as hungarian_assigner * fix init * reformat docstring * Avoid division by zero in PAA head when num_pos=0 * fix cpu (#3948) * add mode for iou_calculator and make giou cost as a default case * make mode as a param in iou_calculator * reformat docsting * make iou_mode outside of iou_calculator Co-authored-by: Cao Yuhang <yhcao6@gmail.com> Co-authored-by: yuzhj <31239926+yuzhj@users.noreply.github.com> Co-authored-by: Wenwei Zhang <40779233+ZwwWayne@users.noreply.github.com> Co-authored-by: mmeendez8 <miguelmndez@gmail.com> Co-authored-by: Yosuke Shinya <42844407+shinya7y@users.noreply.github.com> Co-authored-by: LabMemNo003 <LabMemNo003@gmail.com> Co-authored-by: Ye Liu <yeliudev@outlook.com> Co-authored-by: aboettcher <aboettcher@users.noreply.github.com> Co-authored-by: Ryan Li <xlide@connect.ust.hk> Co-authored-by: Jiangmiao Pang <pangjiangmiao@gmail.com> * Re-implements RandomCrop to support different crop_type (#4093) * re-implements RandomCrop to support different crop_type * add assertion for relative related crop_type * reformat, and add image shape assertion for unit test * reformat * Supports for DETR training mode in process. (#3963) * add input_img_shape in img_meta, and supports training mode toy * toy commit * toy commit * re-organize and support for detr training mode * add unit test for detr training, testing, and transformer_head * add more unit test for transformer * reformat * remove unnecessary * use small transformer model for unit test * reformat docstring * reformat * reformat * Supports DETR e150 config (#4197) * rename e300 config as e150 config * reformat * add comments on override option in Resize * add comments on override option in Resize * position embeddingto positional encoding * fix unit tests * fix registry name bug * rename file * remove unnecessary Co-authored-by: ZwwWayne <wayne.zw@outlook.com> Co-authored-by: Cao Yuhang <yhcao6@gmail.com> Co-authored-by: yuzhj <31239926+yuzhj@users.noreply.github.com> Co-authored-by: Wenwei Zhang <40779233+ZwwWayne@users.noreply.github.com> Co-authored-by: mmeendez8 <miguelmndez@gmail.com> Co-authored-by: Yosuke Shinya <42844407+shinya7y@users.noreply.github.com> Co-authored-by: LabMemNo003 <LabMemNo003@gmail.com> Co-authored-by: Ye Liu <yeliudev@outlook.com> Co-authored-by: aboettcher <aboettcher@users.noreply.github.com> Co-authored-by: Ryan Li <xlide@connect.ust.hk> Co-authored-by: Jiangmiao Pang <pangjiangmiao@gmail.com>

* tag 'v2.5.0': (102 commits) Bump to v2.5.0 (open-mmlab#3879) [Fix]: fix mask rcnn training stuck problem when there is no positive rois (open-mmlab#3713) Add missing notes in data customization (open-mmlab#3906) support to use pytorch 1.6 in docker (open-mmlab#3905) [enhance]: Improve documentation of modules and dataset customization (open-mmlab#3821) fix cv2 import error of ligGL.so.1 (open-mmlab#3891) fix the API change bug of PAA (open-mmlab#3883) Fix typo in bbox_flip (open-mmlab#3886) [Enhance]: Convert mask to bool before using it as img's index for robustness and speedup (open-mmlab#3870) [Docs] Remove duplicate content in docs/config.md (open-mmlab#3875) [Docs] Fix typo in docs/tutorials/new_dataset.md (open-mmlab#3876) Support TTA of ATSS, FCOS, YOLOv3 (open-mmlab#3844) Fix nonzero in NMS for PyTorch 1.6.0 (open-mmlab#3867) [Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images (open-mmlab#3695) fix rpn transforming bug in two stage networks (open-mmlab#3754) Clean background_labels in the dense heads (open-mmlab#3221) improve the function of simple_test_bboxes (open-mmlab#3853) Add doc of modify loss (open-mmlab#3777) fix sabl validating bug (open-mmlab#3849) YOLACT (open-mmlab#3456) ...

mmeendez8 and others added 4 commits June 26, 2020 12:16

Add regression test and test data

c1b0e18

Add fix for empty gt images

65fbf74

Trigger CI build

995f79b

Merge branch 'fix_coco_loader_empty_images' of github.com:Gradiant/mm…

8646888

…detection into fix_coco_loader_empty_images Conflicts: .gitignore tests/test_dataset.py

ZwwWayne changed the title ~~[Fix] dataloader training with empty-GT images~~ [Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images Sep 5, 2020

refactor get_subset_by_classes

cdd100b

ZwwWayne requested a review from Johnson-Wang September 5, 2020 06:47

reformat and fix docstring

887cee8

ZwwWayne mentioned this pull request Sep 5, 2020

Fix coco loader empty images #3133

Closed

[refactor]: move get_subset_by_classes to _filter_imgs

f1d1441

ZwwWayne added 2 commits September 5, 2020 17:34

make img_ids consistent

89c8d23

Merge branch 'master' of github.com:open-mmlab/mmdetection into fix_c…

05784ce

…oco_loader_empty_images

Johnson-Wang reviewed Sep 7, 2020

View reviewed changes

mmdet/datasets/coco.py Show resolved Hide resolved

mmdet/datasets/custom.py Outdated Show resolved Hide resolved

ZwwWayne requested a review from OceanPang September 8, 2020 10:46

resolve comments

654cdc3

Johnson-Wang approved these changes Sep 8, 2020

View reviewed changes

simplify logics

fc8d50c

OceanPang reviewed Sep 9, 2020

View reviewed changes

mmdet/datasets/coco.py Show resolved Hide resolved

mmdet/datasets/custom.py Show resolved Hide resolved

ZwwWayne mentioned this pull request Sep 16, 2020

only call get_subset_by_classes during training #3744

Closed

ZwwWayne requested a review from hellock September 18, 2020 09:43

ZwwWayne requested review from Johnson-Wang and OceanPang September 26, 2020 07:57

Merge branch 'master' of github.com:open-mmlab/mmdetection into fix_c…

baef701

…oco_loader_empty_images

ZwwWayne mentioned this pull request Sep 29, 2020

KeyError: On training model using cascade_rcnn_hrnetv2p_w32_20e_coco.py #3841

Closed

OceanPang approved these changes Sep 29, 2020

View reviewed changes

ZwwWayne added 3 commits September 30, 2020 23:34

add warning

2111bfe

add warning

88db12b

add warning

96715cd

ZwwWayne merged commit b347bf2 into open-mmlab:master Sep 30, 2020

ZwwWayne mentioned this pull request Oct 10, 2020

Unexpected result : On training the model using cascade_rcnn and gfl. #3890

Closed

AndreaPi mentioned this pull request Oct 1, 2021

How to disable empty GT filtering in a custom dataset of type COCODataset #6215

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images #3695

[Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images #3695

ZwwWayne commented Sep 5, 2020 •

edited

Loading

codecov bot commented Sep 5, 2020 •

edited

Loading

hellock commented Sep 30, 2020

[Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images #3695

[Refactor] refactor get_subset_by_classes in dataloader for training with empty-GT images #3695

Conversation

ZwwWayne commented Sep 5, 2020 • edited Loading

codecov bot commented Sep 5, 2020 • edited Loading

Codecov Report

hellock commented Sep 30, 2020

ZwwWayne commented Sep 5, 2020 •

edited

Loading

codecov bot commented Sep 5, 2020 •

edited

Loading