summaryrefslogtreecommitdiff
path: root/include
AgeCommit message (Expand)AuthorFilesLines
2014-02-26Define split layer (merge trick)Evan Shelhamer1-0/+21
2014-02-26fix split layer insertion bug with in-place layersJeff Donahue1-3/+9
2014-02-26get_split_blob_name returns a string to remove some verbosityJeff Donahue1-2/+1
2014-02-26eliminate redundant code with get_split_blob_name methodJeff Donahue1-0/+3
2014-02-26add split layer insertion tests; move split insertion code to util fileJeff Donahue1-0/+21
2014-02-26Added tanh activation function layer.Aravindh Mahendran1-0/+17
2014-02-17Revert "Merge pull request #114 from jeffdonahue/dags-by-split"Evan Shelhamer2-44/+0
2014-02-16get_split_blob_name returns a string to remove some verbosityJeff Donahue1-2/+1
2014-02-15eliminate redundant code with get_split_blob_name methodJeff Donahue1-0/+3
2014-02-15add split layer insertion tests; move split insertion code to util fileJeff Donahue2-3/+21
2014-02-15add SplitLayer and Net::AddSplits to transform shared bottom blobs intoJeff Donahue2-0/+24
2014-02-13added sigmoid layerTobias Domhan1-0/+19
2014-01-31Merge pull request #62 from viirya/masterEvan Shelhamer1-1/+6
2014-01-31Fix test_data_layer segfault by adding destructor to join pthreadKai Li1-0/+1
2014-01-29choose proper thread number per block according to CUDA architecture.Liang-Chi Hsieh1-1/+6
2014-01-19cleanup whitespaceEvan Shelhamer1-1/+1
2013-12-04fix really stupid bug in flatten layer (and add test that shows theJeff Donahue1-1/+1
2013-11-27add flatten layerJeff Donahue1-0/+21
2013-11-21remove remaining distributed solver stuffYangqing Jia1-1/+0
2013-11-21removed the unsuccessful distributed solver codeYangqing Jia1-62/+0
2013-11-19added explanation notes to syncedmemYangqing Jia1-27/+13
2013-11-15Made a major change: when initializing a network, the input size are no longe...Yangqing Jia1-7/+3
2013-11-13pycaffe updateYangqing Jia1-5/+17
2013-11-11io.hpp bugfixYangqing Jia1-1/+1
2013-11-11added infogain loss layerYangqing Jia1-0/+23
2013-11-11bugfixYangqing Jia1-0/+2
2013-11-08misc updateYangqing Jia3-3/+23
2013-11-07distributed solver still having bugs. Pausing for now...Yangqing Jia1-1/+4
2013-11-06distributed solver: small fix, still bugsYangqing Jia1-1/+2
2013-11-06distributed server update. bug in synchronous connections.Yangqing Jia2-1/+2
2013-11-06working asynchronous sgd code. may have errors.Yangqing Jia1-0/+1
2013-11-04distributed solver - not checked thoroughlyYangqing Jia1-0/+57
2013-11-04solver minor changeYangqing Jia1-3/+3
2013-10-31solver restructuring: now all prototxt are specified in the solver protocol b...Yangqing Jia2-4/+10
2013-10-30syncedmem: added code to not use cudamalloc/cudafreeYangqing Jia1-1/+20
2013-10-30pooling layer: added stochastic pooling layer, not tested, compilableYangqing Jia1-0/+1
2013-10-28syncedmem bugfixYangqing Jia1-2/+2
2013-10-28option to run on cpu without gpuYangqing Jia2-1/+26
2013-10-26added back opencv dependency and convert scriptsYangqing Jia1-0/+1
2013-10-25regarding https://plus.google.com/113952791760990667476/posts/Q5fwQY6JeEq - s...Yangqing Jia1-1/+3
2013-10-25bnllYangqing Jia1-0/+19
2013-10-24reverted the layer register effort. Live today, fight tomorrow.Yangqing Jia3-88/+3
2013-10-24put registration commands to cpp filesYangqing Jia2-15/+2
2013-10-24Merge branch 'master' of github.com:Yangqing/caffeYangqing Jia2-0/+11
2013-10-24Register pattern for layersYangqing Jia3-4/+102
2013-10-24common: added DeviceQuery() functionYangqing Jia1-0/+2
2013-10-23need backward computation, and train_net resume point. Not debugged.Yangqing Jia1-0/+6
2013-10-22allow setting custom weight decayYangqing Jia1-0/+3
2013-10-21unchecked: using mean file.Yangqing Jia1-0/+1
2013-10-17cleaning codesYangqing Jia1-0/+4
generated by cgit v1.2.3 (git 2.25.1) at 2024-04-28 11:55:21 +0000