Commits

Breadcrumbs

History for

onmaster

User selector

All users

All time

Commit History

Commits on Nov 18, 2014

• L1Cost CUNN implementation

  ajtulloch

  committed
  ba23c90

  Copy full SHA for ba23c90

Commits on Nov 13, 2014

• Dropout tests

  nicholas-leonard

  committed
  675ef50

  Copy full SHA for 675ef50

Commits on Nov 11, 2014

• fixing global variables in tests

  soumith

  committed
  b269e95

  Copy full SHA for b269e95

Commits on Oct 6, 2014

• SpatialMaxPooling can handle (dW != kW) or (dH ~= kH)

  nicholas-leonard

  committed
  87d55bf

  Copy full SHA for 87d55bf

Commits on Aug 1, 2014

• Incorporated rectangular filter support recently added in Caffe repository.

  jzthree

  committed
  1b9f16a

  Copy full SHA for 1b9f16a

Commits on Jul 22, 2014

• Deprecated SpatialConvolution.

  clementfarabet

  committed
  2741b7c

  Copy full SHA for 2741b7c

• SpatialConvolutionMM now supports arbitrary batch sizes (and 3-dim

  Show description for 629bd1e
  clementfarabet

  committed
  629bd1e

  Copy full SHA for 629bd1e

• Merge pull request #23 from jonathantompson/kl_divergence

  Show description for b6f892b
  soumith

  committed
  b6f892b

  Copy full SHA for b6f892b

• added a CUDA version of Min (basically copy-paste from Max).

  jonathantompson

  committed
  1023bda

  Copy full SHA for 1023bda

• removed redundant kernels from DistKLDivCriterion and modified test.

  jonathantompson

  committed
  927f209

  Copy full SHA for 927f209

Commits on Jul 21, 2014

• added DistKLDivCriterion CUDA implementation and a unit test for it. Also added modified the MSE unit test to include sizeAverage tests as well (to get all code paths).

  jonathantompson

  committed
  95f7a57

  Copy full SHA for 95f7a57

• SpatialConvolutionMM supports dim=3 (once gradBias is implementeD).

  clementfarabet

  committed
  2f67641

  Copy full SHA for 2f67641

• Removed half-baked (and old) module SpatialConvolutionMap

  clementfarabet

  committed
  b4f9078

  Copy full SHA for b4f9078

Commits on Jul 20, 2014

• fixed default tensor in test bug

  nicholas-leonard

  committed
  68a8d67

  Copy full SHA for 68a8d67

Commits on Jul 19, 2014

• harmonized unit test with torch7/cutorch/nn

  nicholas-leonard

  committed
  305022f

  Copy full SHA for 305022f

Commits on Jul 18, 2014

• extended benchmark

  clementfarabet

  committed
  3515a9d

  Copy full SHA for 3515a9d

• typo in test.lua

  soumith

  committed
  c868e82

  Copy full SHA for c868e82

Commits on Jul 17, 2014

• Merge branch 'spatialconvmm' of github.com:torch/cunn into spatialconvmm

  clementfarabet

  committed
  56534b0

  Copy full SHA for 56534b0

• benchmarking accGradParameters()

  clementfarabet

  committed
  dfde301

  Copy full SHA for dfde301

• adding another small case for benchmark

  soumith

  committed
  616dd75

  Copy full SHA for 616dd75

• SpatialConvolutionMM:updateGradInput() tested ok.

  clementfarabet

  committed
  aeebc8f

  Copy full SHA for aeebc8f

Commits on Jul 16, 2014

• added transposed conv (BHWD)

  clementfarabet

  committed
  4c1e718

  Copy full SHA for 4c1e718

Commits on Jul 15, 2014

• typo in benchmark...

  clementfarabet

  committed
  dda0b9a

  Copy full SHA for dda0b9a

• added nOutputPlane == 8 check

  soumith

  committed
  0cb6e7f

  Copy full SHA for 0cb6e7f

• SpatialConvolutionMM:forward() passes tests.

  clementfarabet

  committed
  9c83d66

  Copy full SHA for 9c83d66

Commits on Jun 26, 2014

• fixing test.lua stemming from bug introduced in #16

  soumith

  committed
  5fe62b2

  Copy full SHA for 5fe62b2

• Added cuda version for SpatialUpSamplingNearest module

  

  stencilman

  committed
  c6baf0f

  Copy full SHA for c6baf0f

Commits on Jun 25, 2014

• pretty benchmark table

  soumith

  committed
  ef11e1f

  Copy full SHA for ef11e1f

• fixed merge conflicts.

  jonathantompson

  committed
  e6572cc

  Copy full SHA for e6572cc

• added a SoftPlus CUDA module.

  Show description for 9b35dd8
  jonathantompson

  committed
  9b35dd8

  Copy full SHA for 9b35dd8

• added a cuda Exp.

  jonathantompson

  committed
  31e7aaa

  Copy full SHA for 31e7aaa

Commits on Jun 11, 2014

• fixed LogSoftMax (unit tested)

  nicholas-leonard

  committed
  b617318

  Copy full SHA for b617318

• added batch LogSoftMax unit tests

  nicholas-leonard

  committed
  b96f317

  Copy full SHA for b96f317

Commits on May 13, 2014

• fixed TemporalConvolution bugs, added tests

  nicholas-leonard

  committed
  7ba107f

  Copy full SHA for 7ba107f

Commits on May 12, 2014

• TemporalConvolution unit tests

  nicholas-leonard

  committed
  e742740

  Copy full SHA for e742740