index
:
platform/upstream/caffeonacl
accepted/tizen_5.0_unified
accepted/tizen_5.5_unified
accepted/tizen_5.5_unified_mobile_hotfix
accepted/tizen_5.5_unified_wearable_hotfix
accepted/tizen_unified
armcl-v18.11
master
sandbox/daeinki/armcl-v18.08
sandbox/nmerinov/llvm
tizen
tizen_5.0
tizen_5.5
tizen_5.5_mobile_hotfix
tizen_5.5_tv
tizen_5.5_wearable_hotfix
Domain: Machine Learning / ML Framework; Licenses: BSD-2-Clause;
Inki Dae <inki.dae@samsung.com>
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
src
/
caffe
/
solver.cpp
Age
Commit message (
Expand
)
Author
Files
Lines
2017-01-23
Restore can be invoked on rank > 0
Cyprien Noel
1
-1
/
+0
2017-01-06
Switched multi-GPU to NCCL
Cyprien Noel
1
-25
/
+19
2016-11-25
Revert "solver: check and set type to reconcile class and proto"
Evan Shelhamer
1
-12
/
+0
2016-11-21
solver: check and set type to reconcile class and proto
Evan Shelhamer
1
-0
/
+12
2016-02-25
drop Net inputs + Forward with bottoms
Evan Shelhamer
1
-5
/
+3
2016-01-26
Merge pull request #3132 from bwilbertz/LastIterationLoss
Jeff Donahue
1
-13
/
+24
2015-11-26
replace snprintf with a C++98 equivalent
Tea
1
-5
/
+3
2015-11-12
Fix loss of last iteration when average_loss > 1
Benedikt Wilbertz
1
-13
/
+24
2015-10-16
Add automatic upgrade for solver type
Ronghang Hu
1
-1
/
+1
2015-10-16
Split solver code into one file per solver class
Ronghang Hu
1
-811
/
+0
2015-09-18
refine format of switch case in solver
Yan Chen
1
-8
/
+8
2015-09-14
Merge pull request #3049 from seanbell/check-snapshot-prefix
Ronghang Hu
1
-0
/
+19
2015-09-14
Check that the snapshot directory is writeable before starting training
Sean Bell
1
-0
/
+19
2015-09-09
Minor: missing space in string formatting
Sean Bell
1
-1
/
+1
2015-08-22
Add signal handler and early exit/snapshot to Solver.
J Yegerlehner
1
-8
/
+62
2015-08-14
Adam solver
PatWie
1
-0
/
+104
2015-08-13
Merge pull request #2903 from ronghanghu/multi_gpu
Ronghang Hu
1
-21
/
+55
2015-08-12
Data Layers Parallel for Multi-GPU
Ronghang Hu
1
-6
/
+17
2015-08-10
Clean up and modernize AdaDelta code; add learning rate support; add addition...
Matthias Plappert
1
-156
/
+118
2015-08-10
Updated AdaDelta for modern Caffe; reduced iterations on multi-iter tests
Kevin Bache
1
-23
/
+9
2015-08-10
Implement AdaDelta; add test cases; add mnist examples
Mohamed Omran
1
-0
/
+199
2015-08-09
Multi-GPU
Cyprien Noel
1
-17
/
+40
2015-08-09
Use net_->learnable_params() instead of net_->params() in RMSprop
Ronghang Hu
1
-1
/
+1
2015-08-08
Implement RMSProp Solver
Eren Golge
1
-0
/
+76
2015-08-07
Net: add learnable_params_ used by solvers to correctly handle shared params
Jeff Donahue
1
-32
/
+12
2015-08-07
Snapshot model weights/solver state to HDF5 files.
Eric Tzeng
1
-30
/
+134
2015-05-28
directly normalize accumulated gradients
Evan Shelhamer
1
-3
/
+29
2015-05-27
adjust local learning rate and decay according to gradient accumulation
Evan Shelhamer
1
-2
/
+3
2015-05-27
zero-init param diffs in gradient checker
Jonathan L Long
1
-1
/
+1
2015-05-27
zero-init param diffs and accumulate gradients
Jonathan L Long
1
-1
/
+26
2015-05-27
Solver::MakeUpdate() -> Solver::ApplyUpdate
Evan Shelhamer
1
-2
/
+2
2015-05-26
deduplicate decay and local rate in solver updates
Evan Shelhamer
1
-18
/
+5
2015-05-26
Refactor solvers regularization and logging code
Cyprien Noel
1
-293
/
+207
2015-05-14
Merge pull request #2072 from jeffdonahue/final-snapshot-off-by-one
Jon Long
1
-5
/
+8
2015-03-13
Fix for solver issue pointed out by @moskewcz in #1972
Jeff Donahue
1
-1
/
+1
2015-03-08
Increment iter_ before snapshotting, remove +1 logic -- fixes final
Jeff Donahue
1
-5
/
+8
2015-03-03
Blobs are ND arrays (for N not necessarily equals 4).
Jeff Donahue
1
-10
/
+4
2015-02-17
no phase for the solver to orchestrate
Evan Shelhamer
1
-4
/
+0
2015-02-13
Add gradient clipping -- limit L2 norm of parameter gradients
Jeff Donahue
1
-0
/
+27
2015-01-29
Enhancements for debug_info to display more information.
Jeff Donahue
1
-0
/
+1
2015-01-09
improve const-ness of Net
Jonathan L Long
1
-10
/
+13
2015-01-02
break out Step from Solver
Jonathan L Long
1
-35
/
+38
2014-10-17
back-merge
Sergey Karayev
1
-1
/
+1
2014-10-16
Fix init current_step
Sergio
1
-0
/
+1
2014-10-10
Added Multistep, Poly and Sigmoid learning rate decay policies
Sergio
1
-6
/
+31
2014-10-06
correct naming in comment and message about average_loss
Jonathan L Long
1
-1
/
+1
2014-10-06
[fix] check solver prototxt parsing
Jonathan L Long
1
-1
/
+1
2014-10-01
Back-merge branch 'master' into dev (for fix in PR #1203)
Jeff Donahue
1
-4
/
+4
2014-10-01
[fix] solver indexing of output blobs was incorrect for non-singleton
Jeff Donahue
1
-4
/
+4
2014-09-18
fix warning
qipeng
1
-1
/
+1
[next]