Welcome to
mirror list
, hosted at
ThFree Co
, Russian Federation.
github.com/torch/cunn.git - Unnamed repository; edit this file 'description' to name the repository.
index
:
github.com/torch/cunn.git
1.0-0
lvdmaaten-patch-1
master
newtypes
revert-416-half-fixes
revert-421-cudaGLU
revert-430-newCudaGLU
spatfracmax
Unnamed repository; edit this file 'description' to name the repository.
www-data
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
DataParallelTable.lua
Age
Commit message (
Expand
)
Author
2017-02-21
[DPT] Add option for synchronous copy of model to GPUs
Laurens van der Maaten
2016-12-29
Add support for Half tensors to DataParallelTable.
Gregory Chanan
2016-12-16
Add support for Double tensors to DataParallelTable.
Gregory Chanan
2016-12-09
Fix sliceRange for when nElem < splits
Natalia Gimelshein
2016-11-03
more robust support of small batches when input/output dimensions differ
Jean A. Senellart
2016-11-02
test on inputGpu emptiness symmetric in backward and forward
Jean A. Senellart
2016-09-06
Fix issue with flatten parameters in DataParallelTable
Vivek Narayanan
2016-08-24
fix Lua 5.2 compat
Soumith Chintala
2016-04-27
fixes to DataParallelTable for no parameter networks
soumith
2016-04-26
Fix clearState in DataParallelTable
Adam Paszke
2016-04-25
fix clearState for DataParallelTable
Natalia Gimelshein
2016-02-19
Do flattenParameters() only once
Jerry Lin
2016-02-18
Zero gradients on secondary GPUs after accGradParameters
Sam Gross
2016-02-16
Instantiate impl field in older models
Adam Paszke
2016-02-02
Add multi-threading support in DataParallelTable
Sam Gross
2015-12-04
add nccl support to DataParallelTable
ngimel
2015-11-16
Allow move to CUDA from top-level module
Greg Heinrich
2015-10-30
removing syncs from DPT, adding flattenParams
soumith
2015-06-08
DataParallelTable cleanup.
Clement Farabet
2015-05-29
chmod -x
Clement Farabet
2015-03-10
Added DataParallelTable. Added docs.
Jonathan Tompson