JiahuiYu slimmable_networks issues

JiahuiYu / slimmable_networks

Slimmable Networks, AutoSlim, and Beyond, ICLR 2019, and ICCV 2019

Other

911 stars 131 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

fix up the problem in model_profiling when fed 'Modulelist'

#59 13015517713 opened 1 year ago
0
how to run this code with cpu

#58 Chenhait opened 2 years ago
0
How can I evaluate standalone smaller width model?

#57 abhishekaich27 closed 2 years ago
1
the issues of fuction SlimableConv2D

#56 baiyu12345 opened 3 years ago
1
No module named 'mmdet.ops.nms.gpu_nms'

#55 likunnb opened 3 years ago
0
Mobile Implementation

#54 jsiloto opened 3 years ago
0
Slimmable Network using Mixed Precision

#53 pprp opened 3 years ago
0
Slimming to lower width results in poor accuracy on ARM architectures

#52 octavianmm opened 3 years ago
0
Some questions with the code.

#51 tfwang08 closed 3 years ago
0
BN initialization for tracking performance during training

#50 MingxiLi closed 3 years ago
0
US-WDSR

#49 Booooooooooo opened 3 years ago
1
top1 error = 0.99 when testing

#48 YoungDav closed 4 years ago
1
Size mismatch for bn running_mean and running_var, us_mobilenet_v1

#47 octavianmm opened 4 years ago
0
AutoSlim search

#46 HaloTrouvaille opened 4 years ago
1
Would you share Image Super Resolution example model for Universally Slimmable Network training?

#45 conson0214 closed 4 years ago
1
A Question

#44 Youskrpig closed 4 years ago
1
memory increasing during testing

#43 octavianmm closed 4 years ago
1
Run the experiment with distributed mode

#42 hwang595 closed 4 years ago
1
Question about slimming

#41 dada-thu closed 4 years ago
1
reproducing CIFAR10 results for AutoSlim

#40 RudyChin opened 4 years ago
8
Using Cifar-100 to train slimmable_networks

#39 BcomedianC closed 4 years ago
1
why use slimmable_training instead of universally_slimmable_training: True in Autoslim

#38 GYxiaOH closed 4 years ago
7
model profiling

#37 lyn0102 closed 4 years ago
3
Question about arbitrary width in a universally slimmable network.

#36 sseung0703 closed 4 years ago
8
The accuracy decrease sharply after calibration, even on the given Model zoo

#35 ZhanHuaa closed 4 years ago
1
Size mismatch when loading us_mobilenet_v2_calibrated.pt

#34 pawopawo closed 4 years ago
3
Why bn layers do not have weight_decay?

#33 leoozy closed 4 years ago
1
When will you release Greedy Slimming?

#32 apxlwl closed 4 years ago
1
Is num_sample_training=2 enough for training ?

#31 bhheo closed 4 years ago
3
how can i load us_model at a certain ratio

#30 lucaskyle closed 4 years ago
5
us.mobilenet_v2 cifar

#29 lucaskyle closed 4 years ago
5
width_mult = None when training us-Net

#28 lucaskyle closed 4 years ago
3
Calibration with non-uniform widths at test time

#27 MohammadSamragh closed 5 years ago
6
Fix bug of model loading

#26 JiahuiYu closed 5 years ago
0
Size mismatch when loading us_mobilenet_v1_calibrated.pt

#24 bermanmaxim closed 5 years ago
2
Release training code and bn calibration code for US-Nets

#23 JiahuiYu closed 5 years ago
0
US-Net BatchNorm

#22 semin-park closed 5 years ago
6
USNet for Obeject Detection

#21 VectorYoung closed 5 years ago
1
Memory Leak when Training US-Net

#20 VectorYoung closed 5 years ago
1
Sliming Object detection models

#19 abhigoku10 closed 5 years ago
1
Training Speed, S-Net vs. US-Net

#18 VectorYoung closed 5 years ago
1
Training Speed

#17 VectorYoung closed 5 years ago
4
channel configuration of auto-slim mobilenet-v2

#16 twangnh closed 4 years ago
1
weight decay = 0 for BN parameters

#15 wesleywesleyteng closed 5 years ago
5
Reproduced Result

#14 VectorYoung closed 5 years ago
1
is the shufflenet implemented v1 or v2? Thanks.

#13 liwei46 closed 5 years ago
1
How did you combine the hard label loss and the Distillation loss

#12 VectorYoung closed 5 years ago
7
How to retrain slim-MobilenetV2

#11 Zeitzmz closed 5 years ago
5
How to collect post statistics of BN

#10 VectorYoung closed 5 years ago
8
Exciting work! Some questions about universally slimmable networks

#9 TaojiannanYang closed 5 years ago
3