Commits

Breadcrumbs

History for

onmaster

User selector

All users

All time

Commit History

Commits on Dec 2, 2015

• Remove hamming_distance and popcount

  eelstork

  committed
  36bf811

  Copy full SHA for 36bf811

Commits on Jun 3, 2015

• Add LogLayer

  jeffdonahue

  committed
  8c72fe3

  Copy full SHA for 8c72fe3

Commits on Oct 11, 2014

• fix flaky math functions, remove unnecessary instantiations.

  Yangqing

  committed
  fc5ba94

  Copy full SHA for fc5ba94

Commits on Aug 21, 2014

• Merge pull request #956 from longjon/clean-signbit

  Show description for aee9cd3
  shelhamer

  committed
  aee9cd3

  Copy full SHA for aee9cd3

• specialize cpu_strided_dot before instantiation to fix clang++ build

  shelhamer

  committed
  ac64a7b

  Copy full SHA for ac64a7b

• clean up cpu signbit definition

  Show description for 0ac0f20
  longjon

  committed
  0ac0f20

  Copy full SHA for 0ac0f20

Commits on Aug 18, 2014

• add caffe_cpu_strided_dot for strided dot products

  Show description for e957bbc
  longjon

  committed
  e957bbc

  Copy full SHA for e957bbc

Commits on Aug 14, 2014

• Added absolute value layer, useful for implementation of siamese networks!

  Show description for 1991826
  

  Alireza Shafaei

  committed
  1991826

  Copy full SHA for 1991826

Commits on Aug 12, 2014

• Fix caffe/alt_fn lint errors.

  jeffdonahue

  committed
  b2b1ee6

  Copy full SHA for b2b1ee6

Commits on Aug 7, 2014

• LICENSE governs the whole project so strip file headers

  shelhamer

  committed
  b97b88f

  Copy full SHA for b97b88f

Commits on Jul 29, 2014

• fix some namespace with std::signbit

  

  yajiedesign

  authored and

  jeffdonahue

  committed
  41a5637

  Copy full SHA for 41a5637

Commits on Jul 17, 2014

• collect CUDA includes and calls, separate from CPU-only mode, leave out

  Show description for e52d91e
  shelhamer

  committed
  e52d91e

  Copy full SHA for e52d91e

Commits on Jul 10, 2014

• Replace cudaMemcpy with caffe_gpu_memcpy in SyncedMemory per @longjon

  kloudkl

  committed
  904c2ce

  Copy full SHA for 904c2ce

• Check the GPU mode to decide which memcpy to use

  kloudkl

  committed
  4d16ed5

  Copy full SHA for 4d16ed5

• Avoid using cudaMemcpy for memcpy when there is no GPU and CUDA driver

  kloudkl

  committed
  3bd49ba

  Copy full SHA for 3bd49ba

Commits on Jul 4, 2014

• reduce caffe_copy to instantiations, split off caffe_memcpy for void*

  shelhamer

  committed
  6461365

  Copy full SHA for 6461365

• do all caffe_copy() as UVA mem copy, and drop caffe_gpu_copy()

  Show description for aaa9f24
  shelhamer

  committed
  aaa9f24

  Copy full SHA for aaa9f24

Commits on Jun 28, 2014

• Check that pointers are different before copying in caffe_copy and caffe_gpu_copy

  sguada

  committed
  2dd0bc1

  Copy full SHA for 2dd0bc1

Commits on May 25, 2014

• merge caffe_set definitions; define for int as well

  jeffdonahue

  committed
  0a0f9be

  Copy full SHA for 0a0f9be

• make a Blob<unsigned int> and use in dropout layer

  jeffdonahue

  committed
  dcd5743

  Copy full SHA for dcd5743

Commits on Apr 23, 2014

• cleanup data_layer, add prefetch_rng_ field to it and use instead of

  Show description for 99e10ea
  jeffdonahue

  committed
  99e10ea

  Copy full SHA for 99e10ea

Commits on Apr 18, 2014

• pass caffe rng ref into variate_generator constructor instead of having

  Show description for c795bec
  jeffdonahue

  committed
  c795bec

  Copy full SHA for c795bec

Commits on Apr 9, 2014

• make RNG function outputs the last argument per Google C++ style guidelines

  jeffdonahue

  committed
  d002799

  Copy full SHA for d002799

• make RNG function names more similar to other caffe math function names

  jeffdonahue

  committed
  078e0bf

  Copy full SHA for 078e0bf

• call caffe_set_rng at the end of each vRng function to maintain state

  jeffdonahue

  committed
  db4d3b0

  Copy full SHA for db4d3b0

Commits on Apr 8, 2014

• add LRN within map layer and dependencies (eltwise product and power)

  jeffdonahue

  committed
  42c9d66

  Copy full SHA for 42c9d66

Commits on Mar 29, 2014

• Rename caffe_hamming_distance into caffe_cpu_hamming_distance

  kloudkl

  committed
  db7ea5e

  Copy full SHA for db7ea5e

Commits on Mar 23, 2014

• Rename signbit in macros to sgnbit to avoid conflicts with std::signbit

  kloudkl

  committed
  a288d95

  Copy full SHA for a288d95

• Add signbit math func, simplify GPU defs & instantiations with a macro

  kloudkl

  committed
  dc552e0

  Copy full SHA for dc552e0

• Add and test non-in-place scale math functions for CPU and GPU

  kloudkl

  committed
  b1f6eb0

  Copy full SHA for b1f6eb0

• Use macro to simplify element wise cpu math functions

  kloudkl

  committed
  b458b41

  Copy full SHA for b458b41

• Add and test element wise abs math functions for CPU and GPU

  kloudkl

  committed
  ccae3fa

  Copy full SHA for ccae3fa

• Instantiate caffe_cpu_sign for float and double

  kloudkl

  committed
  f634899

  Copy full SHA for f634899

• Add and test sum of absolute values math functions for CPU and GPU

  kloudkl

  committed
  910f312

  Copy full SHA for 910f312

Commits on Mar 22, 2014

• Set copyright to BVLC and contributors.

  Show description for bece205
  shelhamer

  committed
  bece205

  Copy full SHA for bece205