platform/upstream/caffeonacl.git
9 years agoMoved the loop inside PReLUParamBackward to do the reduction inside the kernel
Mausoom Sarkar [Tue, 13 Oct 2015 13:05:32 +0000 (18:35 +0530)]
Moved the loop inside PReLUParamBackward to do the reduction inside the kernel
Now PReLU backward is taking the same time as forward

Code cleanup
Removed unnecessary code
Fixed indent
merge if(channed_shared_)

9 years agoMerge pull request #3162 from kkhoot/fix_mvn
Ronghang Hu [Sun, 11 Oct 2015 16:48:09 +0000 (09:48 -0700)]
Merge pull request #3162 from kkhoot/fix_mvn

Improve numerical stability of variance computation in MVNLayer

9 years agoImprove numerical stability of variance computation in MVNLayer
Kang Kim [Wed, 7 Oct 2015 16:26:25 +0000 (01:26 +0900)]
Improve numerical stability of variance computation in MVNLayer

9 years agoMerge pull request #3170 from sh1r0/pycaffe_io
Jeff Donahue [Sun, 11 Oct 2015 02:39:22 +0000 (19:39 -0700)]
Merge pull request #3170 from sh1r0/pycaffe_io

Remove the 4D constraint of blobproto IO in python

9 years agoRemove the 4D constraint of blobproto IO in python
sh1r0 [Thu, 8 Oct 2015 16:31:05 +0000 (00:31 +0800)]
Remove the 4D constraint of blobproto IO in python

9 years agoMerge pull request #3164 from eerwitt/layer-documentation-sample-fixes
Ronghang Hu [Wed, 7 Oct 2015 19:41:37 +0000 (12:41 -0700)]
Merge pull request #3164 from eerwitt/layer-documentation-sample-fixes

Documentation References ImageNet Sample Location

9 years agofixes BVLC/caffe#3163
e3 [Wed, 7 Oct 2015 18:52:45 +0000 (11:52 -0700)]
fixes BVLC/caffe#3163

9 years agoMerge pull request #3152 from jeffdonahue/silence-backward-fix
Jeff Donahue [Mon, 5 Oct 2015 21:36:11 +0000 (14:36 -0700)]
Merge pull request #3152 from jeffdonahue/silence-backward-fix

SilenceLayer Backward bugfix (fixes #3151)

9 years agoSilenceLayer Backward bugfix (fixes #3151)
Jeff Donahue [Mon, 5 Oct 2015 21:15:08 +0000 (14:15 -0700)]
SilenceLayer Backward bugfix (fixes #3151)

9 years agoMerge pull request #3058 from zoharby/master
Ronghang Hu [Mon, 5 Oct 2015 07:30:10 +0000 (00:30 -0700)]
Merge pull request #3058 from zoharby/master

Add a caffe.io.write_mean function to the MATLAB interface

9 years agoAdd a caffe.io.write_mean function to the MATLAB interface
zoharby [Fri, 11 Sep 2015 12:06:28 +0000 (15:06 +0300)]
Add a caffe.io.write_mean function to the MATLAB interface

Useful for exporting models from MATLAB (e.g. MatConvNet) to Caffe

9 years agoMerge pull request #3133 from kashefy/badge
Evan Shelhamer [Thu, 1 Oct 2015 17:05:32 +0000 (10:05 -0700)]
Merge pull request #3133 from kashefy/badge

add badge for travis build and license

9 years agoadd badge for travis build and license
Youssef Kashef [Thu, 1 Oct 2015 16:20:23 +0000 (18:20 +0200)]
add badge for travis build and license

9 years agoMerge pull request #3069 from timmeinhardt/argmax
Evan Shelhamer [Thu, 1 Oct 2015 00:43:05 +0000 (17:43 -0700)]
Merge pull request #3069 from timmeinhardt/argmax

Add argmax_param "axis" to maximise output along the specified axis

9 years agoMerge pull request #3128 from hjss06/patch-1
Ronghang Hu [Tue, 29 Sep 2015 16:07:44 +0000 (09:07 -0700)]
Merge pull request #3128 from hjss06/patch-1

Fix a typo

9 years agoFix a typo
Yang Song [Tue, 29 Sep 2015 12:07:52 +0000 (20:07 +0800)]
Fix a typo

Fix a typo in the message.

9 years agoMerge pull request #3073 from ronghanghu/consistent-malloc-free
Ronghang Hu [Sat, 26 Sep 2015 18:14:30 +0000 (11:14 -0700)]
Merge pull request #3073 from ronghanghu/consistent-malloc-free

Ensure consistency between memory alloc and free

9 years agoAdd flag on how host memory is allocated
Ronghang Hu [Wed, 16 Sep 2015 20:39:23 +0000 (13:39 -0700)]
Add flag on how host memory is allocated

Add a bool flag to record whether a host memory is allocated using malloc or
cudaMallocHost, and free correspondingly using this flag, instead of depending on Caffe::mode(), which is mutable during runtime.

9 years agoMerge pull request #3032 from ronghanghu/ban-pythonlayer-in-parallel
Evan Shelhamer [Fri, 25 Sep 2015 21:57:17 +0000 (14:57 -0700)]
Merge pull request #3032 from ronghanghu/ban-pythonlayer-in-parallel

Disallow PythonLayer in Multi-GPU training

9 years agoMerge pull request #3118 from ducha-aiki/patch-1
Evan Shelhamer [Fri, 25 Sep 2015 16:23:32 +0000 (09:23 -0700)]
Merge pull request #3118 from ducha-aiki/patch-1

Fix parse_log.sh against "prefetch queue empty" messages

9 years agoImplement ArgMaxLayerTest for axis param
Tim Meinhardt [Tue, 15 Sep 2015 14:57:55 +0000 (16:57 +0200)]
Implement ArgMaxLayerTest for axis param

9 years agoGeneralise ArgMaxLayerTest bottom blob shape
Tim Meinhardt [Tue, 15 Sep 2015 14:57:37 +0000 (16:57 +0200)]
Generalise ArgMaxLayerTest bottom blob shape

9 years agoUpdate ArgMaxLayer documentation for axis param
Tim Meinhardt [Tue, 15 Sep 2015 14:56:45 +0000 (16:56 +0200)]
Update ArgMaxLayer documentation for axis param

9 years agoImplement ArgMaxLayer forward_cpu and reshape for axis param
Tim Meinhardt [Tue, 15 Sep 2015 14:56:16 +0000 (16:56 +0200)]
Implement ArgMaxLayer forward_cpu and reshape for axis param

9 years agoAdd argmax_param axis
Tim Meinhardt [Tue, 15 Sep 2015 14:55:26 +0000 (16:55 +0200)]
Add argmax_param axis

9 years agoFix parse_log.sh against "prefetch queue empty" messages
Dmytro Mishkin [Fri, 25 Sep 2015 07:00:23 +0000 (10:00 +0300)]
Fix parse_log.sh against "prefetch queue empty" messages

9 years agoMerge pull request #3115 from lukeyeager/fix-generate-sample-data
Jeff Donahue [Thu, 24 Sep 2015 21:10:50 +0000 (14:10 -0700)]
Merge pull request #3115 from lukeyeager/fix-generate-sample-data

Fix generate_sample_data.py

9 years agoFix generate_sample_data.py - bug from #2978
Luke Yeager [Thu, 24 Sep 2015 20:33:11 +0000 (13:33 -0700)]
Fix generate_sample_data.py - bug from #2978

9 years agoMerge pull request #2978 from lukeyeager/h5t_integer
Jeff Donahue [Thu, 24 Sep 2015 20:11:05 +0000 (13:11 -0700)]
Merge pull request #2978 from lukeyeager/h5t_integer

Allow H5T_INTEGER in HDF5 files

9 years agoModify HDF5DataLayerTest to test H5T_INTEGER data
Luke Yeager [Wed, 26 Aug 2015 01:58:45 +0000 (18:58 -0700)]
Modify HDF5DataLayerTest to test H5T_INTEGER data

9 years agoAllow H5T_INTEGER in HDF5 files
Luke Yeager [Thu, 24 Sep 2015 19:35:35 +0000 (12:35 -0700)]
Allow H5T_INTEGER in HDF5 files

9 years agoMerge pull request #2951 from dplarson/examples_readme_link
Evan Shelhamer [Thu, 24 Sep 2015 04:17:45 +0000 (21:17 -0700)]
Merge pull request #2951 from dplarson/examples_readme_link

[examples] fix link to feature visualization notebook

9 years agoMerge pull request #3112 from shelhamer/test-reshape-harder
Evan Shelhamer [Thu, 24 Sep 2015 03:38:44 +0000 (20:38 -0700)]
Merge pull request #3112 from shelhamer/test-reshape-harder

[test] Test Reshape more rigorously

9 years ago[test] TestReshape: check that shapes actually change
Evan Shelhamer [Wed, 23 Sep 2015 20:40:24 +0000 (13:40 -0700)]
[test] TestReshape: check that shapes actually change

Check that output spatial shape varies with input shape while the output
num matches the input num.

9 years ago[test] TestReshape: expect instead of check
Evan Shelhamer [Wed, 23 Sep 2015 20:40:16 +0000 (13:40 -0700)]
[test] TestReshape: expect instead of check

9 years ago[test] TestReshape: check small then large
Evan Shelhamer [Wed, 23 Sep 2015 20:38:29 +0000 (13:38 -0700)]
[test] TestReshape: check small then large

checking large then small can mask failure since the smaller shape
memory will fit within the larger shape.

9 years agoMerge pull request #3096 from longjon/fix-conv-reshape
Jon Long [Mon, 21 Sep 2015 23:27:51 +0000 (16:27 -0700)]
Merge pull request #3096 from longjon/fix-conv-reshape

Fix broken conv/deconv reshaping

9 years agoMerge pull request #3093 from shelhamer/check-xcode-clt-version
Evan Shelhamer [Mon, 21 Sep 2015 18:06:03 +0000 (11:06 -0700)]
Merge pull request #3093 from shelhamer/check-xcode-clt-version

[build] check xcode command line tools version >= 6

9 years agofix broken conv/deconv reshaping caused by reading bottom shape in
Jonathan L Long [Sun, 20 Sep 2015 22:31:59 +0000 (15:31 -0700)]
fix broken conv/deconv reshaping caused by reading bottom shape in
LayerSetUp

This also eliminates the extra copying of bottom's shape.

9 years agoMerge pull request #3095 from longjon/fix-deconv-backward
Jeff Donahue [Sun, 20 Sep 2015 22:16:59 +0000 (15:16 -0700)]
Merge pull request #3095 from longjon/fix-deconv-backward

Fix broken DeconvolutionLayer GPU backward caused by ND conv typo

9 years agofix broken DeconvolutionLayer GPU backward caused by typo
Jonathan L Long [Sun, 20 Sep 2015 21:20:28 +0000 (14:20 -0700)]
fix broken DeconvolutionLayer GPU backward caused by typo

9 years ago[build] check xcode command line tools version >= 6
Evan Shelhamer [Sun, 20 Sep 2015 03:38:18 +0000 (20:38 -0700)]
[build] check xcode command line tools version >= 6

future-proof version check for BLAS libraries on OS X
fix #3092

9 years agoMerge pull request #3083 from yanchen036/master
Jon Long [Sat, 19 Sep 2015 02:56:20 +0000 (19:56 -0700)]
Merge pull request #3083 from yanchen036/master

refine format of switch case statement in solver

9 years agoMerge pull request #2049 from jeffdonahue/nd-convolution
Jeff Donahue [Sat, 19 Sep 2015 01:13:05 +0000 (18:13 -0700)]
Merge pull request #2049 from jeffdonahue/nd-convolution

ND convolution with im2col

9 years agoIm2col and Convolution layers support N spatial axes
Jeff Donahue [Thu, 5 Mar 2015 03:30:17 +0000 (19:30 -0800)]
Im2col and Convolution layers support N spatial axes

9 years agoBlob: add SyncedMemory shape accessor for GPU shape access
Jeff Donahue [Thu, 5 Mar 2015 05:31:34 +0000 (21:31 -0800)]
Blob: add SyncedMemory shape accessor for GPU shape access

9 years agocaffe.proto: generalize ConvolutionParameter to N spatial axes
Jeff Donahue [Thu, 5 Mar 2015 03:27:56 +0000 (19:27 -0800)]
caffe.proto: generalize ConvolutionParameter to N spatial axes

9 years agorefine format of switch case in solver
Yan Chen [Fri, 18 Sep 2015 09:02:16 +0000 (17:02 +0800)]
refine format of switch case in solver

9 years agoMerge pull request #3081 from shelhamer/include-io-by-default
Evan Shelhamer [Thu, 17 Sep 2015 23:38:39 +0000 (16:38 -0700)]
Merge pull request #3081 from shelhamer/include-io-by-default

include IO dependencies by default

9 years ago[build] include IO dependencies by default
Evan Shelhamer [Thu, 17 Sep 2015 22:05:12 +0000 (15:05 -0700)]
[build] include IO dependencies by default

keep old behavior by including leveldb, lmdb, and opencv by default

9 years agoMerge pull request #3074 from ronghanghu/show-use-cpu
Ronghang Hu [Thu, 17 Sep 2015 20:26:30 +0000 (13:26 -0700)]
Merge pull request #3074 from ronghanghu/show-use-cpu

Get back 'USE CPU' print for caffe train

9 years agoMerge pull request #2523 from BonsaiAI/separate-io-dependencies
Jeff Donahue [Thu, 17 Sep 2015 07:33:11 +0000 (00:33 -0700)]
Merge pull request #2523 from BonsaiAI/separate-io-dependencies

Separate dependencies for configurable installation

9 years agoAdd a comment indicating that Travis CI tests are CPU only
T.E.A de Souza [Tue, 8 Sep 2015 04:20:40 +0000 (12:20 +0800)]
Add a comment indicating that Travis CI tests are CPU only

9 years agoFix case in CMake notices
Tea [Thu, 17 Sep 2015 07:02:45 +0000 (15:02 +0800)]
Fix case in CMake notices

9 years agoSeparate IO dependencies
Tea [Sat, 27 Jun 2015 03:44:56 +0000 (11:44 +0800)]
Separate IO dependencies

OpenCV, LMDB, LevelDB and Snappy are made optional via switches
(USE_OPENCV, USE_LMDB, USE_LEVELDB) available for Make and CMake
builds. Since Snappy is a LevelDB dependency, its use is determined by
USE_LEVELDB. HDF5 is left bundled because it is used for serializing
weights and solverstates.

9 years agoGet back 'USE CPU' print for caffe train
Ronghang Hu [Wed, 16 Sep 2015 19:06:16 +0000 (12:06 -0700)]
Get back 'USE CPU' print for caffe train

9 years agoMerge pull request #3070 from mohomran/resize_nd-image_bug
Jeff Donahue [Wed, 16 Sep 2015 18:24:27 +0000 (11:24 -0700)]
Merge pull request #3070 from mohomran/resize_nd-image_bug

bug in caffe.io.resize_image when applied to images with 2 or >3 channels

9 years agoremoved bug in caffe.io.resize_image when applied to Nd images
Mohamed Omran [Tue, 15 Sep 2015 15:18:32 +0000 (17:18 +0200)]
removed bug in caffe.io.resize_image when applied to Nd images

9 years agoMerge pull request #3049 from seanbell/check-snapshot-prefix
Ronghang Hu [Mon, 14 Sep 2015 21:57:18 +0000 (14:57 -0700)]
Merge pull request #3049 from seanbell/check-snapshot-prefix

Check that the snapshot directory is writeable before starting training

9 years agoCheck that the snapshot directory is writeable before starting training
Sean Bell [Wed, 9 Sep 2015 16:49:27 +0000 (12:49 -0400)]
Check that the snapshot directory is writeable before starting training

9 years agoMerge pull request #3063 from ronghanghu/use-expectnear-in-eltwiselayertest
Ronghang Hu [Mon, 14 Sep 2015 17:14:56 +0000 (10:14 -0700)]
Merge pull request #3063 from ronghanghu/use-expectnear-in-eltwiselayertest

Use EXPECT_NEAR in EltwiseLayer test

9 years agoMerge pull request #2941 from thatguymike/MultiGPUDocs
Ronghang Hu [Mon, 14 Sep 2015 16:54:35 +0000 (09:54 -0700)]
Merge pull request #2941 from thatguymike/MultiGPUDocs

Add some documentation on Multi-GPU support

9 years agoUse EXPECT_NEAR in EltwiseLayer test
Ronghang Hu [Mon, 14 Sep 2015 03:46:24 +0000 (20:46 -0700)]
Use EXPECT_NEAR in EltwiseLayer test

Otherwise there seem to be some numerical issues causing BLAS results not exactly same as evaluated results in test code.

9 years agoMerge pull request #3048 from seanbell/minor-string-fix
Ronghang Hu [Wed, 9 Sep 2015 17:25:08 +0000 (10:25 -0700)]
Merge pull request #3048 from seanbell/minor-string-fix

Minor: missing space in string formatting

9 years agoMinor: missing space in string formatting
Sean Bell [Wed, 9 Sep 2015 16:52:52 +0000 (12:52 -0400)]
Minor: missing space in string formatting

9 years agodisallow PythonLayer in Multi-GPU training
Ronghang Hu [Fri, 4 Sep 2015 21:16:31 +0000 (14:16 -0700)]
disallow PythonLayer in Multi-GPU training

9 years agoMerge pull request #3024 from danielgordon10/python-solver-fix
Evan Shelhamer [Fri, 4 Sep 2015 17:36:31 +0000 (14:36 -0300)]
Merge pull request #3024 from danielgordon10/python-solver-fix

[pycaffe] expose all solvers for direct instantiation (although note get_solver)

9 years agoenabling the alternate solvers to be accessed by the python interface
Daniel Gordon [Fri, 4 Sep 2015 17:12:00 +0000 (10:12 -0700)]
enabling the alternate solvers to be accessed by the python interface

9 years agoMerge pull request #3026 from CDLuminate/patch-1
Jeff Donahue [Fri, 4 Sep 2015 05:50:13 +0000 (22:50 -0700)]
Merge pull request #3026 from CDLuminate/patch-1

Fix a typo

9 years agoMerge pull request #3027 from ronghanghu/fix-accuracy-test
Evan Shelhamer [Fri, 4 Sep 2015 05:47:51 +0000 (02:47 -0300)]
Merge pull request #3027 from ronghanghu/fix-accuracy-test

Fix AccuracyLayerTest for per-class accuracy.

9 years agoFix AccuracyLayerTest for per-class accuracy.
Ronghang Hu [Fri, 4 Sep 2015 04:44:45 +0000 (21:44 -0700)]
Fix AccuracyLayerTest for per-class accuracy.

Fix AccuracyLayerTest for per-class accuracy. Previously in #2935, it crashes since the test accuracy is nan (0/0) when a class never appear.

9 years agoUpdate extract_features.cpp
Lumin Zhou [Fri, 4 Sep 2015 04:38:43 +0000 (04:38 +0000)]
Update extract_features.cpp

9 years agoMerge pull request #3023 from jeffdonahue/param-debug-info-fix
Evan Shelhamer [Fri, 4 Sep 2015 00:14:40 +0000 (21:14 -0300)]
Merge pull request #3023 from jeffdonahue/param-debug-info-fix

fix Net debug_info to report learnable params

9 years agoMerge pull request #2959 from jeffdonahue/netspec-allow-non-iterable-repeated
Jeff Donahue [Thu, 3 Sep 2015 23:51:07 +0000 (16:51 -0700)]
Merge pull request #2959 from jeffdonahue/netspec-allow-non-iterable-repeated

NetSpec: don't require lists to specify single-element repeated fields

9 years agoNetSpec: don't require lists to specify single-element repeated fields
Jeff Donahue [Sat, 22 Aug 2015 00:29:06 +0000 (17:29 -0700)]
NetSpec: don't require lists to specify single-element repeated fields

9 years agonet.cpp fix debug_info params -> learnable_params
Jeff Donahue [Thu, 3 Sep 2015 21:57:19 +0000 (14:57 -0700)]
net.cpp fix debug_info params -> learnable_params

9 years agoMerge pull request #3014 from jeffdonahue/trivial-slice-concat
Ronghang Hu [Thu, 3 Sep 2015 20:08:30 +0000 (13:08 -0700)]
Merge pull request #3014 from jeffdonahue/trivial-slice-concat

Allow trivial slice/concat

9 years agoConcatLayer: allow trivial operation with single bottom Blob
Jeff Donahue [Fri, 23 Jan 2015 20:52:44 +0000 (12:52 -0800)]
ConcatLayer: allow trivial operation with single bottom Blob

9 years agoSliceLayer: allow trivial operation with single top Blob
Jeff Donahue [Tue, 7 Oct 2014 18:55:54 +0000 (11:55 -0700)]
SliceLayer: allow trivial operation with single top Blob

9 years agoMerge pull request #2990 from mattdawkins/add-openblas-path
Jeff Donahue [Wed, 2 Sep 2015 02:37:41 +0000 (19:37 -0700)]
Merge pull request #2990 from mattdawkins/add-openblas-path

Add extra OpenBLAS include search path

9 years agoMerge pull request #3010 from darrengarvey/fixup-memory-leak-siamese
Jeff Donahue [Wed, 2 Sep 2015 00:55:03 +0000 (17:55 -0700)]
Merge pull request #3010 from darrengarvey/fixup-memory-leak-siamese

Fix memory leak in convert_mnist_siamese_data.

9 years agoMerge pull request #3012 from lukeyeager/convert-imageset-logging
Jeff Donahue [Wed, 2 Sep 2015 00:43:28 +0000 (17:43 -0700)]
Merge pull request #3012 from lukeyeager/convert-imageset-logging

Show output from convert_imageset tool

9 years agoMerge pull request #3011 from darrengarvey/fixup-cifar10-example
Jeff Donahue [Wed, 2 Sep 2015 00:42:26 +0000 (17:42 -0700)]
Merge pull request #3011 from darrengarvey/fixup-cifar10-example

cifar10: Fix examples by setting snapshot_format.

9 years agoMerge pull request #3009 from darrengarvey/cleanup-documentation-warnings
Jeff Donahue [Wed, 2 Sep 2015 00:41:05 +0000 (17:41 -0700)]
Merge pull request #3009 from darrengarvey/cleanup-documentation-warnings

Fix up documentation errors.

9 years agoShow output from convert_imageset tool
Luke Yeager [Wed, 2 Sep 2015 00:20:37 +0000 (17:20 -0700)]
Show output from convert_imageset tool

9 years agoFix up documentation errors.
Darren Garvey [Tue, 1 Sep 2015 23:54:06 +0000 (00:54 +0100)]
Fix up documentation errors.

Fix some doxygen warnings about an undocumented argument in Blob and
incorrect documentation for SoftmaxWithLossLayer::Forward_cpu().

9 years agocifar10: Fix examples by setting snapshot_format.
Darren Garvey [Tue, 1 Sep 2015 19:09:25 +0000 (20:09 +0100)]
cifar10: Fix examples by setting snapshot_format.

Commit 4227828a set the default binary format from HDF5 to BINARYPROTO to
fix #2885. This broke the cifar10 examples which relied on this default.

This commit specifies the snapshot_format explicitly since the rest of the
example relies on this being HDF5.

9 years agoMerge pull request #3008 from darrengarvey/cleanup-cmake-variable
Jeff Donahue [Tue, 1 Sep 2015 23:25:35 +0000 (16:25 -0700)]
Merge pull request #3008 from darrengarvey/cleanup-cmake-variable

Cleanup: Fixup capitalisation of Caffe_POSTFIX.

9 years agoFix memory leak in convert_mnist_siamese_data.
Darren Garvey [Mon, 10 Aug 2015 01:16:20 +0000 (02:16 +0100)]
Fix memory leak in convert_mnist_siamese_data.

This fixes a memory leak by using delete[] rather than plain delete.

9 years agoCleanup: Fixup capitalisation of Caffe_POSTFIX.
Darren Garvey [Tue, 1 Sep 2015 14:42:26 +0000 (15:42 +0100)]
Cleanup: Fixup capitalisation of Caffe_POSTFIX.

Replaces CAffe_POSTFIX -> Caffe_POSTFIX.

9 years agoMerge pull request #3007 from philkr/neg_lr_mult
Jeff Donahue [Tue, 1 Sep 2015 20:42:15 +0000 (13:42 -0700)]
Merge pull request #3007 from philkr/neg_lr_mult

Compute backward for negative lr_mult

9 years agoCompute backward for negative lr_mult
philkr [Tue, 1 Sep 2015 20:11:26 +0000 (13:11 -0700)]
Compute backward for negative lr_mult

9 years agoMerge pull request #2998 from longjon/data-race
Jeff Donahue [Sun, 30 Aug 2015 20:43:20 +0000 (13:43 -0700)]
Merge pull request #2998 from longjon/data-race

Fix a recently introduced race condition in DataLayer

9 years agoMerge pull request #2894 from mfigurnov/fix-truncation-warning
Ronghang Hu [Sun, 30 Aug 2015 05:40:12 +0000 (22:40 -0700)]
Merge pull request #2894 from mfigurnov/fix-truncation-warning

Fix truncation of value warning

9 years agofix GPU data race
Jonathan L Long [Sat, 29 Aug 2015 04:27:11 +0000 (21:27 -0700)]
fix GPU data race

Previously, the prefetch GPU -> top GPU and prefetch CPU -> prefetch GPU
copies were launched concurrently in separate streams, allowing the next
batch to be copied in before the current one is read.

This patch explicitly synchronizes the prefetch -> top copy wrt the
host, preventing the CPU -> GPU from being launched until its
completion.

9 years agoMerge pull request #2989 from jyegerlehner/embed-layer-compile-warning
Jeff Donahue [Thu, 27 Aug 2015 16:30:00 +0000 (09:30 -0700)]
Merge pull request #2989 from jyegerlehner/embed-layer-compile-warning

Fix EmbedLayer compiler warning for unused variable.

9 years agoFix EmbedLayer compiler warning for unused variable.
J Yegerlehner [Thu, 27 Aug 2015 15:47:14 +0000 (10:47 -0500)]
Fix EmbedLayer compiler warning for unused variable.

9 years agoAdd extra openblas search path
Matt Dawkins [Thu, 27 Aug 2015 14:51:36 +0000 (10:51 -0400)]
Add extra openblas search path

9 years agoMerge pull request #2944 from philkr/python_layer_param
Jon Long [Thu, 27 Aug 2015 01:12:17 +0000 (18:12 -0700)]
Merge pull request #2944 from philkr/python_layer_param

Give the python layer parameter/weight blobs.

9 years agoMerge pull request #2970 from ronghanghu/spp-fix
Ronghang Hu [Wed, 26 Aug 2015 19:20:32 +0000 (12:20 -0700)]
Merge pull request #2970 from ronghanghu/spp-fix

Fix SPPLayer top blob num and address `pyramid_height_ == 1`