Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / PeizeSun/SparseR-CNN issues and pull requests

#99 - How can I use a custom dataset?

Issue - State: open - Opened by JEO-96 almost 3 years ago - 2 comments

#97 - How to install the Sparse R-CNN(it seems amusing)?

Issue - State: open - Opened by xiaomin0416 about 3 years ago - 1 comment

#97 - How to install the Sparse R-CNN(it seems amusing)?

Issue - State: open - Opened by xiaomin0416 about 3 years ago - 1 comment

#96 - Confusing about the performance in the paper

Issue - State: closed - Opened by YJHMITWEB about 3 years ago - 3 comments

#96 - Confusing about the performance in the paper

Issue - State: closed - Opened by YJHMITWEB about 3 years ago - 3 comments

#95 - How to do use test time augmentation?

Issue - State: open - Opened by bravezzzzzz about 3 years ago - 1 comment

#95 - How to do use test time augmentation?

Issue - State: open - Opened by bravezzzzzz about 3 years ago - 1 comment

#94 - Change ADAMW to SGD

Issue - State: closed - Opened by mxin262 over 3 years ago - 4 comments

#94 - Change ADAMW to SGD

Issue - State: closed - Opened by mxin262 over 3 years ago - 4 comments

#93 - Does anything changes in detectron2？

Issue - State: open - Opened by kuiran over 3 years ago - 1 comment

#93 - Does anything changes in detectron2？

Issue - State: open - Opened by kuiran over 3 years ago - 1 comment

#92 - Question about the proposal_boxes in the code

Issue - State: closed - Opened by botaoye over 3 years ago - 2 comments

#92 - Question about the proposal_boxes in the code

Issue - State: closed - Opened by botaoye over 3 years ago - 2 comments

#91 - Resume started from zero again!

Issue - State: open - Opened by zobeirraisi over 3 years ago - 3 comments

#91 - Resume started from zero again!

Issue - State: open - Opened by zobeirraisi over 3 years ago - 3 comments

#90 - How the feature re-use is performed

Issue - State: open - Opened by XuyangBai over 3 years ago - 1 comment

#90 - How the feature re-use is performed

Issue - State: open - Opened by XuyangBai over 3 years ago - 1 comment

#89 - 'assert (boxes1[:, 2:] >= boxes1[:, :2]).all()' happened when training

Issue - State: open - Opened by Swall0w over 3 years ago - 4 comments

#89 - 'assert (boxes1[:, 2:] >= boxes1[:, :2]).all()' happened when training

Issue - State: open - Opened by Swall0w over 3 years ago - 4 comments

#88 - how to visualize train log?

Issue - State: closed - Opened by Yixiang-Wu over 3 years ago

#88 - how to visualize train log?

Issue - State: closed - Opened by Yixiang-Wu over 3 years ago

#87 - how to use plot function?

Issue - State: open - Opened by Yixiang-Wu over 3 years ago - 1 comment

#86 - Question about the evaluation

Issue - State: closed - Opened by FL77N over 3 years ago - 10 comments

#86 - Question about the evaluation

Issue - State: closed - Opened by FL77N over 3 years ago - 10 comments

#85 - Init proposal box Visualization

Issue - State: closed - Opened by junha1125 over 3 years ago - 2 comments

#85 - Init proposal box Visualization

Issue - State: closed - Opened by junha1125 over 3 years ago - 2 comments

#84 - Question about the Iterative architecture

Issue - State: open - Opened by GuangxingHan over 3 years ago - 2 comments

#83 - About the loss format

Issue - State: open - Opened by FL77N over 3 years ago - 4 comments

#82 - train w/o pre-trained backbone

Issue - State: open - Opened by StarsMyDestination over 3 years ago - 3 comments

#81 - question about pred_bboxes.detach() in DynamicHead.forward()

Issue - State: open - Opened by burningisnaive over 3 years ago - 2 comments

#81 - question about pred_bboxes.detach() in DynamicHead.forward()

Issue - State: open - Opened by burningisnaive over 3 years ago - 2 comments

#80 - the learning rate adjustment when training with one gpu or two gpus

Issue - State: open - Opened by Young0111 over 3 years ago - 5 comments

#80 - the learning rate adjustment when training with one gpu or two gpus

Issue - State: open - Opened by Young0111 over 3 years ago - 5 comments

#79 - 保存模型与提供模型大小之间的差异

Issue - State: closed - Opened by luyao-cv over 3 years ago - 1 comment

#78 - Multi-head attention in dynamic head

Issue - State: closed - Opened by HYUNJS over 3 years ago - 3 comments

#78 - Multi-head attention in dynamic head

Issue - State: closed - Opened by HYUNJS over 3 years ago - 3 comments

#77 - Question regarding proposal feature

Issue - State: open - Opened by abeyang00 over 3 years ago - 4 comments

#77 - Question regarding proposal feature

Issue - State: open - Opened by abeyang00 over 3 years ago - 4 comments

#76 - Create __init__.py

Pull Request - State: closed - Opened by dashu233 over 3 years ago

#75 - Hi,Have you verified the model to ONNX, i meet some bugs to onnx.

Issue - State: open - Opened by csliuchang over 3 years ago - 2 comments

#75 - Hi,Have you verified the model to ONNX, i meet some bugs to onnx.

Issue - State: open - Opened by csliuchang over 3 years ago - 2 comments

#74 - Does input data annotations need to be normalized coordinate before training?

Issue - State: closed - Opened by sonnguyenasu over 3 years ago - 2 comments

#74 - Does input data annotations need to be normalized coordinate before training?

Issue - State: closed - Opened by sonnguyenasu over 3 years ago - 2 comments

#73 - Will you release DetCo code and model version?

Issue - State: closed - Opened by Mingfeng-Wang over 3 years ago - 3 comments

#73 - Will you release DetCo code and model version?

Issue - State: closed - Opened by Mingfeng-Wang over 3 years ago - 3 comments

#72 - issues about Mix-Precision training

Issue - State: open - Opened by changewOw over 3 years ago - 1 comment

#71 - how to do Grid and Rondom proposal boxes initialization?

Issue - State: open - Opened by liz6688 over 3 years ago - 5 comments

#71 - how to do Grid and Rondom proposal boxes initialization?

Issue - State: open - Opened by liz6688 over 3 years ago - 5 comments

#70 - Detco

Pull Request - State: closed - Opened by iFighting over 3 years ago

#69 - detco pretrain support

Pull Request - State: closed - Opened by iFighting over 3 years ago

#68 - When set num_heads = 1, doesnt train

Issue - State: open - Opened by abeyang00 over 3 years ago - 12 comments

#68 - When set num_heads = 1, doesnt train

Issue - State: open - Opened by abeyang00 over 3 years ago - 12 comments

#67 - cuda out of memory issue

Issue - State: closed - Opened by abeyang00 almost 4 years ago

#67 - cuda out of memory issue

Issue - State: closed - Opened by abeyang00 almost 4 years ago

#66 - Have you tried smooth l1 + giou?

Issue - State: closed - Opened by Mingfeng-Wang almost 4 years ago - 1 comment

#65 - question about learned proposal boxes

Issue - State: open - Opened by liyunsheng13 almost 4 years ago - 1 comment

#65 - question about learned proposal boxes

Issue - State: open - Opened by liyunsheng13 almost 4 years ago - 1 comment

#63 - robust focal loss when alpah=-1

Pull Request - State: closed - Opened by ShoufaChen almost 4 years ago

#63 - robust focal loss when alpah=-1

Pull Request - State: closed - Opened by ShoufaChen almost 4 years ago

#62 - Randomness in the results for COCO

Issue - State: closed - Opened by AyeshaNirma almost 4 years ago - 2 comments

#62 - Randomness in the results for COCO

Issue - State: closed - Opened by AyeshaNirma almost 4 years ago - 2 comments

#61 - after trained about 7000 iters.it has error.

Issue - State: open - Opened by henbucuoshanghai almost 4 years ago - 2 comments

#61 - after trained about 7000 iters.it has error.

Issue - State: open - Opened by henbucuoshanghai almost 4 years ago - 2 comments

#60 - how to change the confidence score in train net.py when eval-only?

Issue - State: open - Opened by yeloiss almost 4 years ago - 1 comment

#60 - how to change the confidence score in train net.py when eval-only?

Issue - State: open - Opened by yeloiss almost 4 years ago - 1 comment

#59 - how to detect mang images one time?

Issue - State: closed - Opened by henbucuoshanghai almost 4 years ago - 1 comment

#58 - how to detect mang pictures one time?

Issue - State: open - Opened by henbucuoshanghai almost 4 years ago - 1 comment

#56 - the ap is higher than 1,why?

Issue - State: closed - Opened by henbucuoshanghai almost 4 years ago - 2 comments

#56 - the ap is higher than 1,why?

Issue - State: closed - Opened by henbucuoshanghai almost 4 years ago - 2 comments

#54 - question about value of base learning rate

Issue - State: closed - Opened by liyunsheng13 almost 4 years ago - 2 comments

#52 - An error was encountered while the demo was running

Issue - State: closed - Opened by Baboom-l almost 4 years ago - 1 comment

#51 - Cannot reproduce the results for Res50

Issue - State: closed - Opened by liyunsheng13 almost 4 years ago - 4 comments

#49 - 如何负采样及负样例的loss在box那里怎么算？

Issue - State: closed - Opened by Zssssss almost 4 years ago

#49 - 如何负采样及负样例的loss在box那里怎么算？

Issue - State: closed - Opened by Zssssss almost 4 years ago

#48 - Can you add a simple tutorial about custom dataset?

Issue - State: closed - Opened by wen0618 almost 4 years ago - 2 comments

#48 - Can you add a simple tutorial about custom dataset?

Issue - State: closed - Opened by wen0618 almost 4 years ago - 2 comments

#47 - How can I check the learned bboxes?

Issue - State: closed - Opened by Mingfeng-Wang almost 4 years ago - 2 comments

#47 - How can I check the learned bboxes?

Issue - State: closed - Opened by Mingfeng-Wang almost 4 years ago - 2 comments

#46 - 请问怎样训练PASCAL VOC2007 ，VOC2012呢？ tks

Issue - State: closed - Opened by henbucuoshanghai almost 4 years ago - 3 comments

#45 - pos_cost_class and neg_cost_class

Issue - State: open - Opened by abcxs almost 4 years ago - 1 comment

#43 - how to resume training

Issue - State: closed - Opened by freedom521jin almost 4 years ago - 2 comments

#43 - how to resume training

Issue - State: closed - Opened by freedom521jin almost 4 years ago - 2 comments

#42 - 用rrpn的方式训练SparseRCNN

Issue - State: closed - Opened by shehuimao almost 4 years ago - 1 comment

#42 - 用rrpn的方式训练SparseRCNN

Issue - State: closed - Opened by shehuimao almost 4 years ago - 1 comment

#41 - Train code used is different with demo.py code

Issue - State: closed - Opened by Mingfeng-Wang almost 4 years ago

#40 - How can change the code used in demo.py?

Issue - State: closed - Opened by Mingfeng-Wang almost 4 years ago

#39 - Should use 100 proposals for evaluation for fair comparisons with other method.

Issue - State: open - Opened by bearcatt almost 4 years ago - 2 comments

#39 - Should use 100 proposals for evaluation for fair comparisons with other method.

Issue - State: open - Opened by bearcatt almost 4 years ago - 2 comments