Welcome to
mirror list
, hosted at
ThFree Co
, Russian Federation.
github.com/soumith/cudnn.torch.git - Unnamed repository; edit this file 'description' to name the repository.
index
:
github.com/soumith/cudnn.torch.git
R1
R2
R3
R4
R5
R6
R7
functional-findex
half
master
revert-231-algo
volfullconv
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
/
SpatialConvolution.lua
Age
Commit message (
Expand
)
Author
2016-11-10
Improved existing 16->32 fallback. Added performance-based fallback.
Boris Fomitchev
2016-11-02
clearState in FindEx
Sergey Zagoruyko
2016-10-19
Added new refactoring for convolution and filter descriptors
Boris Fomitchev
2016-10-19
Code review changes
Boris Fomitchev
2016-09-26
Fixed sticky algo modes
Boris Fomitchev
2016-09-23
Revamped workspace handling in find.lua
Boris Fomitchev
2016-08-30
Addressed code review comments
Boris Fomitchev
2016-08-23
Stream awareness restored. Better WS encapsulation
Boris Fomitchev
2016-08-21
FindEx implementetation + refactoring, take 3
Boris Fomitchev
2016-08-06
working double precision
soumith
2016-08-06
Revert "Refactoring CUDNN Find"
revert-231-algo
Soumith Chintala
2016-08-04
Completing cudnnFind refactoring; addressing code review notes
Boris Fomitchev
2016-08-03
Refactoring cudnnFind
Boris Fomitchev
2016-08-02
Fixed algo location
Boris Fomitchev
2016-08-02
Refactoring CUDNN Find
Boris Fomitchev
2016-07-08
handle half true/pseudo
Sergey Zagoruyko
2016-06-23
half, double with tests
Sergey Zagoruyko
2016-06-23
forward fp16 convolution works
Sergey Zagoruyko
2016-04-20
move gradInput resize to updateGradInput
Sergey Zagoruyko
2016-04-13
R5 rebase
Sergey Zagoruyko
2016-04-08
use nn.utils.clear
Sergey Zagoruyko
2016-03-30
move gradInput resize to updateGradInput
Sergey Zagoruyko
2016-03-22
check if padW and padH exist
Sergey Zagoruyko
2016-03-17
adding _input and _gradOutput to clearState
soumith
2016-02-26
allow non-contiguous inputs for Spatial and Volumetric convolution
soumith
2016-02-16
removing streams optimization
soumith
2016-02-12
clearState
Sergey Zagoruyko
2016-01-26
adding :noBias() to SpatialConvolution
soumith
2016-01-26
cudnn<>nn conversion
Sergey Zagoruyko
2016-01-26
syncing uptil master/R3 9bc3cbac4f054438f5e77824f868cd94e9e22f81
soumith
2016-01-11
Make bias optional in cudnn.SpatialConvolution
Sam Gross
2015-11-06
integrating changes from master
soumith
2015-10-14
Conmpleting the merge
Boris Fomitchev
2015-10-14
Fixing conflicts
Boris Fomitchev
2015-10-14
Bugfix by Andrei
Boris Fomitchev
2015-10-09
Merge pull request #64 from borisfom/master
Soumith Chintala
2015-10-07
cudnn.h fully synced for V4. addTensor_v2 uses changed to new AddTensor API.
Boris Fomitchev
2015-10-05
Merge pull request #62 from borisfom/master
Soumith Chintala
2015-10-03
Initial cudnn_v4 compatibility patch. cudnn4 Batch Normaliztion Support added...
Boris Fomitchev
2015-10-02
fix typo SpatialConvolution with groups
Alban Desmaison
2015-10-01
adding caching for autotuner results
soumith
2015-09-15
fastest mode disabled in Spatial conv
soumith
2015-09-15
adding proper documentation for new modes
soumith
2015-09-15
add back stuff that went missing in merge
soumith
2015-09-15
whitespace cleanups, fixing logsoftmax test
soumith
2015-09-15
fix as suggested in #44
soumith
2015-08-25
small fixes
soumith
2015-08-23
flag to enable or not to enable auto-tuner
soumith
2015-08-21
fix for cudnn r3 perf bug
soumith
2015-08-20
shared workspace
soumith
[next]