vacancy/Synchronized-BatchNorm-PyTorch
Synchronized Batch Normalization implementation in PyTorch.
PythonMIT
Issues
- 3
test gap between training and test
#30 opened by ZhiweiYan-96 - 0
- 0
.
#49 opened by LSKhappychild - 0
how to export with onnx
#48 opened by tongchangD - 1
- 7
- 2
raining couldn 't start
#45 opened by zwyking - 4
Is it more difficult to train model?
#18 opened by yuanfuqiang456 - 7
Training cannot start
#40 opened by shuuchen - 4
Train Stucked
#44 opened by YinengXiong - 4
a question about the highlight "use sqrt(max(var, eps)) instead of sqrt(var + eps)"
#43 opened by louyj136 - 10
is everyone test the code in pytorch 1.0 ?
#24 opened by wadesunyang - 9
- 2
Is this a bug that channel between input tensor and sync batchnorm are mismatch the code still run successful?
#31 opened by lscelory - 0
module问题
#39 opened by fanjing136 - 1
batchnor while using distributed dataparallel
#42 opened by ParthaEth - 0
Training cannot start
#41 opened by shuuchen - 2
Question on `sqrt(max(var, eps))`
#38 opened by nmhkahn - 1
How to use it when testing
#33 opened by zhulanyun - 1
Wired things, module must have its parameters and buffers on device cuda:0 (device_ids[0]) but found one of them on device: cpu
#37 opened by kehuantiantang - 1
- 2
- 0
How to cite this repo in bib?
#32 opened by Anikily - 18
- 1
Where is the "BatchNormReimpl"?
#29 opened by skanderbug - 2
Network performance is getting worse
#28 opened by charlesCXK - 6
about convert_model
#26 opened by Re3write - 9
Wired things happen when applied to FPN
#3 opened by ShuLiu1993 - 2
- 1
Different from torch.nn.DataParallel that will separate total batch to each gpu equally?
#23 opened by ksnzh - 1
Can one past the test ?
#17 opened by chenchr - 1
shall I use your `DataParallelWithCallback` instead of pytorch `nn.DataParallel`?
#19 opened by CoinCheung - 1
About python test scripts and SyncBN usage
#21 opened by Spritea - 5
about fp16
#22 opened by 666zz666 - 6
- 7
Another test script for numerical stability
#14 opened by acgtyrant - 2
- 2
- 1
How to solve the numeric error?
#12 opened by acgtyrant - 1
How to load bias weights
#7 opened by guantinglin - 1
The mean is not averaged realy.
#4 opened by acgtyrant - 3
How to use this module to load the pre-trained model offered by the official PyTorch?
#6 opened by ycszen - 1