Age | Commit message (Expand) | Author | Files | Lines |
2017-08-26 | add support acl batch normal,direct conv, local connect, concat layers | honggui | 7 | -51/+362 |
2017-06-02 | 1. Porting Caffe onto ARM Compute Library. | Yao Honggui | 17 | -3/+968 |
2017-03-07 | Merge pull request #4630 from BlGene/load_hdf5_fix | Evan Shelhamer | 1 | -2/+2 |
2017-01-18 | Merge pull request #5098 from yaronli/master | Evan Shelhamer | 1 | -1/+4 |
2017-01-17 | Merge pull request #4563 from cypof/nccl | Evan Shelhamer | 15 | -233/+166 |
2017-01-06 | Using default from proto for prefetch | Cyprien Noel | 1 | -3/+0 |
2017-01-06 | Python layers should build on multiprocess & solver_cnt; enable with bindings | Marian Gläser | 1 | -1/+1 |
2017-01-06 | Switched multi-GPU to NCCL | Cyprien Noel | 15 | -231/+167 |
2016-12-21 | Use mkl_malloc when use mkl | Tomasz Socha | 1 | -0/+12 |
2016-12-16 | check leveldb iterator status for snappy format. | liyangguang | 1 | -1/+4 |
2016-11-25 | Revert "solver: check and set type to reconcile class and proto" | Evan Shelhamer | 1 | -2/+0 |
2016-11-21 | solver: check and set type to reconcile class and proto | Evan Shelhamer | 1 | -0/+2 |
2016-11-16 | sigmoid cross-entropy loss: normalize loss by different schemes | Evan Shelhamer | 1 | -0/+11 |
2016-11-15 | sigmoid cross-entropy loss: ignore selected targets by `ignore_label` | Evan Shelhamer | 1 | -0/+5 |
2016-11-01 | corrected typo in accuracy_layer.hpp: MaxTopBlos -> MaxTopBlobs | baecchi | 1 | -1/+1 |
2016-10-27 | sigmoid cross-entropy loss: add GPU forward for full GPU mode | Evan Shelhamer | 1 | -0/+2 |
2016-10-22 | Fix: made load_hd5 check blob dims by default. | max argus | 1 | -2/+2 |
2016-09-12 | batch norm: auto-upgrade old layer definitions w/ param messages | Evan Shelhamer | 1 | -0/+6 |
2016-09-12 | batch norm: hide statistics from solver, simplifying layer definition | Evan Shelhamer | 1 | -4/+2 |
2016-09-12 | [docs] identify batch norm layer blobs | Evan Shelhamer | 1 | -11/+12 |
2016-09-09 | [docs] clarify handling of bias and scaling by BiasLayer, ScaleLayer | Evan Shelhamer | 3 | -15/+15 |
2016-08-29 | Merge pull request #4647 from ClimbsRocks/patch-3 | Jeff Donahue | 1 | -1/+1 |
2016-08-29 | Merge pull request #4646 from ClimbsRocks/patch-2 | Jeff Donahue | 1 | -1/+1 |
2016-08-28 | changes "c++" to "C++" for consistency | Preston Parry | 1 | -1/+1 |
2016-08-28 | fixes typo- duplicate "a a" | Preston Parry | 1 | -1/+1 |
2016-08-28 | updates tense in docs | Preston Parry | 1 | -1/+1 |
2016-08-18 | Merge pull request #3272 from ixartz/master | Evan Shelhamer | 1 | -0/+5 |
2016-06-03 | Add level and stages to Net constructor | Luke Yeager | 1 | -0/+1 |
2016-06-01 | Add LSTMLayer and LSTMUnitLayer, with tests | Jeff Donahue | 1 | -0/+154 |
2016-06-01 | Add RNNLayer, with tests | Jeff Donahue | 1 | -0/+47 |
2016-06-01 | Add RecurrentLayer: an abstract superclass for other recurrent layer types | Jeff Donahue | 1 | -0/+187 |
2016-05-16 | Add cuDNN v5 support, drop cuDNN v3 support | Felix Abecassis | 4 | -3/+24 |
2016-05-04 | add parameter layer for learning any bottom | Jonathan L Long | 1 | -0/+45 |
2016-05-04 | Merge pull request #3995 from ZhouYzzz/python-phase | Jon Long | 1 | -0/+1 |
2016-04-20 | Don't set map_size=1TB in util/db_lmdb | Luke Yeager | 1 | -5/+8 |
2016-04-15 | Allow the python layer have attribute "phase" | ZhouYzzz | 1 | -0/+1 |
2016-04-14 | CropLayer: groom comments | Evan Shelhamer | 1 | -0/+9 |
2016-03-05 | Merge pull request #3590 from junshi15/GPUUtilities | Jon Long | 1 | -0/+5 |
2016-03-05 | Merge pull request #3588 from junshi15/P2psyncPrepare | Jon Long | 1 | -1/+4 |
2016-03-05 | split p2psync::run() | Jun Shi | 1 | -1/+4 |
2016-03-05 | Crop: fixes, tests and negative axis indexing. | max argus | 1 | -2/+2 |
2016-03-05 | Extend Crop to N-D, changed CropParameter. | max argus | 1 | -2/+20 |
2016-03-05 | add CropLayer: crop blob to another blob's dimensions with offsets | Jonathan L Long | 1 | -0/+49 |
2016-03-04 | add check and find GPU device utilities | Jun Shi | 1 | -0/+5 |
2016-02-26 | Deprecate ForwardPrefilled(), Forward(bottom, loss) in lieu of dropping | Evan Shelhamer | 1 | -0/+9 |
2016-02-25 | collect Net inputs from Input layers | Evan Shelhamer | 1 | -2/+11 |
2016-02-25 | drop Net inputs + Forward with bottoms | Evan Shelhamer | 1 | -27/+7 |
2016-02-25 | deprecate input fields and upgrade automagically | Evan Shelhamer | 1 | -0/+6 |
2016-02-25 | add InputLayer for Net input | Evan Shelhamer | 1 | -0/+44 |
2016-02-20 | tranpose parameter added to IP layer to support tied weights in an autoencode... | Youssef Kashef | 1 | -0/+1 |