KimSoybean ScratchDet issues - Githubissues

KimSoybean / ScratchDet

The code and models for paper: "ScratchDet: Exploring to Train Single-Shot Object Detectors from Scratch"

229 stars 38 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Welcome update to OpenMMLab 2.0

#16 vansin opened 1 year ago
1
对于fasterrcn

#15 zhaoyanjoin opened 3 years ago
1
Could you provide the weights

#14 shengtao96 opened 4 years ago
1
The training occurs NAN and -inf, anyone knows what happened here?Iteration 20, loss = -nan I0905 19:18:20.666388 10119 solver.cpp:259] Train net output #0: mbox_loss = inf (* 1 = inf loss) I0905 19:18:20.666396 10119 sgd_solver.cpp:138] Iteration 20, lr = 0.05

#13 xjiao004 opened 5 years ago
0
Some questions

#12 herrickli opened 5 years ago
0
there is no function 'weighted_smoothl1 ' in mmdet.core?

#11 jshuzi closed 5 years ago
1
About the location of BN in the head of SSD VGG-16

#10 LaVieEnRoseSMZ opened 5 years ago
0
How to train ScratchDet with mmdetection？

#9 yanlongbinluck opened 5 years ago
9
@foralliance Hi！I think we should speak English because your questions may help the people in other countries.

#8 defei-coder opened 5 years ago
3
How do I train my data set?

#7 anjingxing opened 5 years ago
1
关于Pytorch模型转换过来的VGG-16-BN

#6 xncaffe opened 5 years ago
1
Will you or did you try Group Normalization

#5 XCRobert opened 5 years ago
1
train VGG-SSD from scratch get a better result

#4 ujsyehao opened 5 years ago
1
about some details

#3 foralliance opened 5 years ago
4
How soon will the code and model be here？

#2 feng150416 closed 5 years ago
2
Add code and models

#1 impredicative opened 5 years ago
14