summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Expand)AuthorFilesLines
2015-08-15Malloc at least 1 byte for MultiGPU P2PSync buffersRonghang Hu1-1/+2
2015-08-14Fix MultiGPU solver test with TEST_GPUID != 1Ronghang Hu1-2/+13
2015-08-14Merge pull request #2921 from buaaliyi/multi_gpuRonghang Hu1-2/+7
2015-08-14Malloc at least one byte in ParallelRonghang Hu1-1/+3
2015-08-14Merge pull request #2918 from ronghanghu/adamRonghang Hu3-28/+253
2015-08-14Adam solverPatWie3-28/+253
2015-08-14Destroy CUDA stream when finishedLI Yi1-2/+7
2015-08-13Merge pull request #2572 from flx42/optimize_inner_product_special_caseJeff Donahue2-8/+54
2015-08-13Merge pull request #2903 from ronghanghu/multi_gpuRonghang Hu20-230/+1240
2015-08-12Apply mutex only to shared layers and fix NVCC warningRonghang Hu2-3/+31
2015-08-12Data Layers Parallel for Multi-GPURonghang Hu4-12/+50
2015-08-11fix for learnable_param_ids_Jeff Donahue1-0/+1
2015-08-10Clean up and modernize AdaDelta code; add learning rate support; add addition...Matthias Plappert2-234/+251
2015-08-10Updated AdaDelta for modern Caffe; reduced iterations on multi-iter testsKevin Bache2-25/+11
2015-08-10Implement AdaDelta; add test cases; add mnist examplesMohamed Omran3-2/+298
2015-08-09Detect topology corner cases and improve broadcast ordermhouston1-32/+41
2015-08-09Multi-GPUCyprien Noel10-106/+695
2015-08-09Add DataReader for parallel training with one DB sessionCyprien Noel7-37/+174
2015-08-09Persistent prefetch threadCyprien Noel8-69/+129
2015-08-09Change the way threads are started and stoppedCyprien Noel3-26/+27
2015-08-09Thread-local CaffeCyprien Noel3-4/+66
2015-08-09Add BlockingQueue for inter-thread communicationCyprien Noel1-0/+86
2015-08-09Merge pull request #2887 from shelhamer/solver-test-dataEvan Shelhamer4-24/+45
2015-08-09Use net_->learnable_params() instead of net_->params() in RMSpropRonghang Hu1-1/+1
2015-08-09Encapsulate kRMSDecay in solver testsRonghang Hu1-98/+58
2015-08-08Implement RMSProp SolverEren Golge3-48/+298
2015-08-08TestGradientBasedSolver: replace dummy data with hdf5Evan Shelhamer4-23/+45
2015-08-08TestGradientBasedSolver: drop doubled seed inititializationEvan Shelhamer1-1/+0
2015-08-07temporarily switch the snapshot_format default back to BINARYPROTOJeff Donahue2-2/+2
2015-08-07Net: add learnable_params_ used by solvers to correctly handle shared paramsJeff Donahue4-98/+241
2015-08-07TestGradientBasedSolver: make tests across solver types more consistentJeff Donahue1-32/+59
2015-08-07TestGradientBasedSolver: restore Gaussian filler to all tests exceptJeff Donahue1-2/+7
2015-08-07TestSnapshot expects .h5 snapshots, explicitly checks history.Eric Tzeng1-1/+24
2015-08-07Snapshot model weights/solver state to HDF5 files.Eric Tzeng9-109/+424
2015-08-07TestGradientBasedSolver: add TestSnapshot to verify behavior whenJeff Donahue1-5/+107
2015-08-07add double_data, double_diff to BlobProto for weights/snapshots savedJeff Donahue2-7/+44
2015-08-06PythonLayer takes parameters by stringTakuya Narihira1-0/+5
2015-08-04ImageData layer default batch size of 1, and check for zero batch sizephilkr2-3/+6
2015-08-03Change log levels in upgrade_protoLuke Yeager1-10/+10
2015-07-20Merge pull request #2740 from philkr/travis_python3Evan Shelhamer1-0/+5
2015-07-20Travis scripts for python3 and pytest for cmake. Also fixes CUDA CMake build ...philkr1-0/+5
2015-07-10Removes a unused variable warningphilkr1-1/+0
2015-07-02Optimize inner product layer for special case M == 1Felix Abecassis2-8/+54
2015-06-23Merge pull request #2611 from dpaiton/test_net_bugfixEvan Shelhamer1-4/+7
2015-06-22fixed _force_color check, fixes #2635berleon1-2/+2
2015-06-16fixed two bugs with prototext formatDylan Paiton1-4/+7
2015-06-05Split db.hpp into leveldb_db.hpp and lmdb_db.hppSergio Guadarrama3-55/+74
2015-06-04LogLayer gpu functionality moved to .cu fileMalte Stær Nissen2-47/+57
2015-06-03[bug] fix double instantiation of GPU methods in LogLayerEvan Shelhamer1-2/+0
2015-06-02Add ReductionLayer to reduce any number of "tail" axes to a scalar valueJeff Donahue4-1/+554