Search results

训练速度过慢

Status: Open.

#83 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· bonnie-cbw opened on Jun 28, 2021

Gradient about IOU-Smooth L1 loss

Status: Open.

#82 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· igo312 opened on Jun 10, 2021

cuda报错

Status: Open.

#81 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· kk-letsgo opened on May 18, 2021

请问iou_rotate.py里面的这个利用gpu计算iou的rbbx_overlaps是否限制只能输入tensorflow的tensor?

Status: Open.

#80 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· tjmannn opened on Dec 23, 2020

How to calculate mAP when training on DOTA dataset? Should I write the code myself?

Status: Open.

#79 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· Lanxin1011 opened on Dec 3, 2020

FPN中RoI Pooling层的公式疑惑

Status: Open.

#78 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· dexterod opened on Dec 2, 2020

使用cfgs_resnet50_dota_v4和cfgs_resnet50_dota_v1失函数一直震荡，

Status: Open.

#74 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· zsh1234567 opened on Sep 13, 2020

采用配置文件cfgs_res50_dota_v5（即v1 + iou-smooth l1 loss）cls loss始终在1-1.2之间，reg loss 出现nan

Status: Open.

#71 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· mingmingzhu123 opened on Jun 29, 2020

Do I need to design how to calulate Map?

Status: Open.

#65 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· colorfullywonderfully opened on Mar 13, 2020

How to choose MobilenetV2 as backbone?

Status: Open.

#63 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· hunaodaotianming opened on Mar 2, 2020

training and test UCAS-AOD dataset

enhancementNew feature or requestNew feature or request

good first issueGood for newcomersGood for newcomers

Status: Open.

#57 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· liusiyuan1111 opened on Dec 28, 2019

erro with nms_rotate_gpu

Status: Open.

#55 In DetectionTeamUCAS/RetinaNet_Tensorflow_Rotation;

· ZOUYIyi opened on Dec 24, 2019