Ecosyste.ms: Issues

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

GitHub / facebookresearch/adaptive_teacher issues and pull requests

#72 - gtfine problem

Issue - State: open - Opened by PlutoNov about 1 year ago

#68 - Selecting custom classes from dataset

Issue - State: open - Opened by Manjuphoenix over 1 year ago

#66 - demo

Issue - State: open - Opened by wansd over 1 year ago - 1 comment

#64 - Save teacher weight

Issue - State: open - Opened by ShenZheng2000 over 1 year ago - 1 comment

#62 - How to train a SourceOnly model or Oracle model?

Issue - State: closed - Opened by Donghuan777 over 1 year ago - 2 comments

#60 - How

Issue - State: closed - Opened by Donghuan777 over 1 year ago

#59 - How to get the AP of each label?

Issue - State: closed - Opened by judyxt over 1 year ago - 2 comments

#58 - backbone modification

Issue - State: open - Opened by achintski over 1 year ago - 1 comment

#57 - FloatingPointError: Predicted boxes or scores contain Inf/NaN

Issue - State: closed - Opened by anranbixin over 1 year ago - 9 comments

#56 - The Beta parameter of Foggy Cityscapes

Issue - State: open - Opened by lch1999 over 1 year ago - 1 comment

#55 - TypeError: register_buffer() takes 3 positional arguments but 4 were given

Issue - State: open - Opened by HuanrongLIU over 1 year ago - 3 comments

#54 - How to train with the coco format datasets?

Issue - State: closed - Opened by Donghuan777 over 1 year ago - 1 comment

#53 - How to open the output file

Issue - State: open - Opened by fatherplus over 1 year ago - 1 comment

#52 - what is crop ratio?

Issue - State: open - Opened by victor00070 almost 2 years ago - 1 comment

#51 - question about selecting best model (validation or post last step?)

Issue - State: open - Opened by shjustinbaek almost 2 years ago - 1 comment

#50 - [fix #49] Invalid use of function `align_and_update_state_dicts`

Pull Request - State: closed - Opened by hnanacc almost 2 years ago - 3 comments
Labels: CLA Signed

#48 - resnet backbone dont converge

Issue - State: open - Opened by victor00070 almost 2 years ago - 1 comment

#47 - how to change backbone to resnet

Issue - State: closed - Opened by victor00070 almost 2 years ago - 1 comment

#46 - Very low performance during evaluation

Issue - State: closed - Opened by victor00070 almost 2 years ago

#45 - Why 2 seperate annotation files?

Issue - State: closed - Opened by victor00070 almost 2 years ago - 4 comments

#44 - Memory Leak

Issue - State: closed - Opened by GeoffreyChen777 about 2 years ago - 1 comment

#42 - mAP predicted by student_model is sometimes higher than teacher

Issue - State: open - Opened by firekeepers about 2 years ago - 7 comments

#41 - why were the mAP and AP50 getting smaller with the increase of iterations

Issue - State: open - Opened by sysuzgg about 2 years ago - 1 comment

#40 - detectron2 v0.6 has removed FastRCNNOutputs class

Issue - State: open - Opened by alvinti about 2 years ago - 1 comment

#39 - Teacher or Student model at inference time?

Issue - State: closed - Opened by cmckee13 about 2 years ago - 1 comment

#38 - About discriminator in the paper

Issue - State: closed - Opened by YangJae96 about 2 years ago - 1 comment

#37 - Why your test scale is larger than 600?

Issue - State: open - Opened by Yorionice1 about 2 years ago - 3 comments

#36 - About VGG16 pre-trained on ImageNet

Issue - State: open - Opened by pengjw23 about 2 years ago - 3 comments

#35 - possible to wrap the teacher model by DistributedDataParallel?

Issue - State: closed - Opened by Weijiang-Xiong about 2 years ago - 2 comments

#34 - Can adaptive teacher use one-stage detector?

Issue - State: closed - Opened by huanmx about 2 years ago - 3 comments

#33 - Can not reproduce the results on "foggy cityscapes" due to Out of Memory Issue

Issue - State: open - Opened by onkarkris about 2 years ago - 7 comments

#31 - The AP and APl is 0.000 when evaluation in cityscapes_foggy_val

Issue - State: open - Opened by Doris1231 about 2 years ago - 1 comment

#30 - AttributeError: 'NoneType' object has no attribute 'keys'

Issue - State: open - Opened by zwang123 over 2 years ago - 6 comments

#29 - Pre-trained VGG16 on ImageNet not shared?

Issue - State: closed - Opened by mecarill over 2 years ago - 1 comment

#28 - The checkpoint state_dict contains keys that are not used by the model

Issue - State: open - Opened by ad45675 over 2 years ago - 3 comments

#27 - AP NaN

Issue - State: closed - Opened by darkhan-s over 2 years ago - 4 comments

#26 - Optimal number of learning iterations

Issue - State: open - Opened by miurakenichi over 2 years ago - 10 comments

#25 - Loss becomes NaN

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

#24 - Can not reproduce the results of "cityscapes" to "foggy cityscapes".

Issue - State: open - Opened by cwhgn over 2 years ago - 12 comments

#23 - Why your DA-FRCNN implementation uses multi-scale training trick?

Issue - State: open - Opened by tmp12316 over 2 years ago - 3 comments

#22 - Evaluation for each category

Issue - State: closed - Opened by helq2612 over 2 years ago - 2 comments

#20 - why ap ap50 ap75 = 0.000?

Issue - State: closed - Opened by oopsboy over 2 years ago - 12 comments

#19 - The pretrained model for VGG16

Issue - State: closed - Opened by ygzhu over 2 years ago - 1 comment

#18 - Poor performance with registered Coco annotations

Issue - State: open - Opened by apexx77 over 2 years ago - 4 comments

#16 - The experiments of vgg16 without BN layers?

Issue - State: open - Opened by kinredon over 2 years ago - 16 comments

#15 - how to get AP50 result for each classes as reported on the paper?

Issue - State: closed - Opened by tyunit over 2 years ago - 9 comments

#14 - loss nan

Issue - State: closed - Opened by Pandaxia8 over 2 years ago - 1 comment

#13 - Why discriminator is trained in supervised and target branch?

Issue - State: closed - Opened by Pandaxia8 over 2 years ago - 2 comments

#12 - How to train using single gpu?

Issue - State: open - Opened by tyunit over 2 years ago - 13 comments

#11 - .

Issue - State: closed - Opened by tyunit over 2 years ago

#10 - .

Issue - State: closed - Opened by tyunit over 2 years ago

#9 - Predicted boxes or scores contain Inf/NaN. Training has diverged.

Issue - State: open - Opened by michaelku1 over 2 years ago - 41 comments

#8 - Regarding the Watercolor Dataset Config

Issue - State: closed - Opened by michaelku1 over 2 years ago - 2 comments

#7 - Fix the bug regarding the parallel distributed training in PyTorch

Pull Request - State: closed - Opened by yujheli over 2 years ago
Labels: CLA Signed

#6 - How to train on custom datasets?

Issue - State: open - Opened by litingfeng over 2 years ago - 9 comments

#5 - Distributed training failure

Issue - State: closed - Opened by litingfeng over 2 years ago - 9 comments

#4 - Question about Figure 4

Issue - State: closed - Opened by tmp12316 over 2 years ago - 2 comments

#2 - Add copyright comment

Pull Request - State: closed - Opened by yujheli over 2 years ago
Labels: CLA Signed

#1 - Code upload by the first author of the paper

Pull Request - State: closed - Opened by yujheli over 2 years ago - 3 comments
Labels: CLA Signed