AgeCommit message (Collapse)AuthorFilesLines
2015-10-01add badge for travis build and licenseYoussef Kashef1-0/+3
2015-10-01Merge pull request #3069 from timmeinhardt/argmaxEvan Shelhamer4-28/+205
Add argmax_param "axis" to maximise output along the specified axis
2015-09-29Merge pull request #3128 from hjss06/patch-1Ronghang Hu1-1/+1
Fix a typo
2015-09-29Fix a typoYang Song1-1/+1
Fix a typo in the message.
2015-09-26Merge pull request #3073 from ronghanghu/consistent-malloc-freeRonghang Hu2-9/+14
Ensure consistency between memory alloc and free
2015-09-25Add flag on how host memory is allocatedRonghang Hu2-9/+14
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.
2015-09-25Merge pull request #3032 from ronghanghu/ban-pythonlayer-in-parallelEvan Shelhamer1-0/+6
Disallow PythonLayer in Multi-GPU training
2015-09-25Merge pull request #3118 from ducha-aiki/patch-1Evan Shelhamer1-1/+6
Fix against "prefetch queue empty" messages
2015-09-25Implement ArgMaxLayerTest for axis paramTim Meinhardt2-13/+140
2015-09-25Generalise ArgMaxLayerTest bottom blob shapeTim Meinhardt1-5/+7
2015-09-25Update ArgMaxLayer documentation for axis paramTim Meinhardt1-3/+9
2015-09-25Implement ArgMaxLayer forward_cpu and reshape for axis paramTim Meinhardt1-15/+38
2015-09-25Add argmax_param axisTim Meinhardt3-5/+24
2015-09-25Fix against "prefetch queue empty" messagesDmytro Mishkin1-1/+6
2015-09-24Merge pull request #3115 from lukeyeager/fix-generate-sample-dataJeff Donahue1-2/+2
2015-09-24Fix - bug from #2978Luke Yeager1-2/+2
2015-09-24Merge pull request #2978 from lukeyeager/h5t_integerJeff Donahue3-7/+36
Allow H5T_INTEGER in HDF5 files
2015-09-24Modify HDF5DataLayerTest to test H5T_INTEGER dataLuke Yeager2-6/+8
2015-09-24Allow H5T_INTEGER in HDF5 filesLuke Yeager1-1/+28
2015-09-24Merge pull request #2951 from dplarson/examples_readme_linkEvan Shelhamer1-1/+1
[examples] fix link to feature visualization notebook
2015-09-24Merge pull request #3112 from shelhamer/test-reshape-harderEvan Shelhamer1-6/+20
[test] Test Reshape more rigorously
2015-09-24[test] TestReshape: check that shapes actually changeEvan Shelhamer1-2/+14
Check that output spatial shape varies with input shape while the output num matches the input num.
2015-09-24[test] TestReshape: expect instead of checkEvan Shelhamer1-2/+2
2015-09-24[test] TestReshape: check small then largeEvan Shelhamer1-2/+4
checking large then small can mask failure since the smaller shape memory will fit within the larger shape.
2015-09-21Merge pull request #3096 from longjon/fix-conv-reshapeJon Long4-16/+11
Fix broken conv/deconv reshaping
2015-09-21Merge pull request #3093 from shelhamer/check-xcode-clt-versionEvan Shelhamer1-2/+3
[build] check xcode command line tools version >= 6
2015-09-20fix broken conv/deconv reshaping caused by reading bottom shape inJonathan L Long4-16/+11
LayerSetUp This also eliminates the extra copying of bottom's shape.
2015-09-20Merge pull request #3095 from longjon/fix-deconv-backwardJeff Donahue1-1/+1
Fix broken DeconvolutionLayer GPU backward caused by ND conv typo
2015-09-20fix broken DeconvolutionLayer GPU backward caused by typoJonathan L Long1-1/+1
2015-09-20[build] check xcode command line tools version >= 6Evan Shelhamer1-2/+3
future-proof version check for BLAS libraries on OS X fix #3092
2015-09-19Merge pull request #3083 from yanchen036/masterJon Long2-15/+15
refine format of switch case statement in solver
2015-09-19Merge pull request #2049 from jeffdonahue/nd-convolutionJeff Donahue21-321/+1594
ND convolution with im2col
2015-09-19Im2col and Convolution layers support N spatial axesJeff Donahue19-311/+1554
2015-09-19Blob: add SyncedMemory shape accessor for GPU shape accessJeff Donahue2-0/+13
2015-09-19caffe.proto: generalize ConvolutionParameter to N spatial axesJeff Donahue1-10/+27
2015-09-18refine format of switch case in solverYan Chen2-15/+15
2015-09-17Merge pull request #3081 from shelhamer/include-io-by-defaultEvan Shelhamer2-5/+10
include IO dependencies by default
2015-09-17[build] include IO dependencies by defaultEvan Shelhamer2-5/+10
keep old behavior by including leveldb, lmdb, and opencv by default
2015-09-17Merge pull request #3074 from ronghanghu/show-use-cpuRonghang Hu1-0/+1
Get back 'USE CPU' print for caffe train
2015-09-17Merge pull request #2523 from BonsaiAI/separate-io-dependenciesJeff Donahue40-61/+266
Separate dependencies for configurable installation
2015-09-17Add a comment indicating that Travis CI tests are CPU onlyT.E.A de Souza1-0/+1
2015-09-17Fix case in CMake noticesTea1-2/+2
2015-09-17Separate IO dependenciesTea40-60/+264
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.
2015-09-16Get back 'USE CPU' print for caffe trainRonghang Hu1-0/+1
2015-09-16Merge pull request #3070 from mohomran/resize_nd-image_bugJeff Donahue1-1/+1
bug in when applied to images with 2 or >3 channels
2015-09-15removed bug in when applied to Nd imagesMohamed Omran1-1/+1
2015-09-14Merge pull request #3049 from seanbell/check-snapshot-prefixRonghang Hu2-0/+21
Check that the snapshot directory is writeable before starting training
2015-09-14Check that the snapshot directory is writeable before starting trainingSean Bell2-0/+21
2015-09-14Merge pull request #3063 from ronghanghu/use-expectnear-in-eltwiselayertestRonghang Hu1-2/+2
Use EXPECT_NEAR in EltwiseLayer test
2015-09-14Merge pull request #2941 from thatguymike/MultiGPUDocsRonghang Hu1-0/+26
Add some documentation on Multi-GPU support