Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Jul 28, 2016

• Examples/SequenceToSequence/CMUDict/Data/cmudict-0.7b.mapping: normalize LE

  mahilleb-msft

  committed
  08691e9

Commits on Jul 27, 2016

• addressed CR feedback

  frankseide

  committed
  540cd0b

• Merge branch 'master' of https://github.com/Microsoft/CNTK into fseide/clonebs

  

  U-FAREAST\fseide

  committed
  6a6ebc4

• fix in Convolution mapCount change

  frankseide

  committed
  da1f9f4

• Makefile: cppevalclient: link to stub libnvidia-ml.so (if real one not present)

  Show description for 24cb169
  mahilleb-msft

  committed
  24cb169

• mapDims parameter to Convolution() is now optional, will be inferred from weight matrix

  frankseide

  committed
  8b68af3

• (fix for previous fix)

  frankseide

  committed
  ee23bb2

• renamed reduction kernels that expect a hard-coded number of threads to reflect that number in their names

  frankseide

  committed
  8f71698

Commits on Jul 26, 2016

• merged from master

  frankseide

  committed
  9dbf806

• undid accidental change of Shuffle()

  frankseide

  committed
  1a80a6a

• addressed minor feedback from Amit's CR;

  Show description for 5e357fe
  frankseide

  committed
  5e357fe

• fix 1BitDGD

  thilow

  committed
  2c1b102

• Integrate thilow/ReduceMax2 into master

  

  Project Philly

  committed
  6a6cff0

• merge with master

  thilow

  committed
  4e4598e

• Added comments

  thilow

  committed
  dcc7e9b

• CNTK v2 library: Add member method to NDMask to enable retreiving the buffer underlying a Mask object

  amitaga

  committed
  759fa18

Commits on Jul 25, 2016

• Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/cntkv2LibraryOperators

  amitaga

  committed
  3e6926d

• Merge branch 'master' of https://github.com/Microsoft/CNTK into amitaga/cntkv2Library

  amitaga

  committed
  575bbf6

• Using double as aggregator, hoping to fix issue with tests TWRGE TWRGS, TLRGS

  thilow

  committed
  524c527

• Removed 'typedef' in partial specialisation of TensorOpReduction in hope to fix Linux build.

  thilow

  committed
  52dda16

• Fixed comments

  thilow

  committed
  24c0ad1

• Renamed variable to reductionOp

  thilow

  committed
  2115db6

• factored aggregation op into a lambda

  thilow

  committed
  5dbb725

• Normalize line endings where missing

  mah

  committed
  44d127e

• CNTK v2 library: Use dynamic loading of CompositeDataReader instead of a static dependency

  amitaga

  committed
  4164856

Commits on Jul 23, 2016

• CNTK v2 library: Expose 18 additional operators in the v2 API

  amitaga

  committed
  a632393

Commits on Jul 22, 2016

• CNTK v2 library: Added initial Trainer and MinibatchSource implementations, enabling end-to-end model training with v2 API. A test added to train a feed-forward network on the MNIST dataset using t…

  Show description for 5b1c217
  amitaga

  committed
  5b1c217

• Merge branch 'master' of https://github.com/Microsoft/CNTK into fseide/clonebs

  

  U-FAREAST\fseide

  committed
  a5d15f3

• bug fix: BatchNormalization::BackpropTo() should distinguish between saveMean and runMean the same way ForwardProp() does

  frankseide

  committed
  7416f9b

• (typo)

  frankseide

  committed
  27ff6f7

• BatchNorm: moved allocation of saveMean() to where they are produced, and allocating them empty when they are not produced at all

  frankseide

  committed
  07a6fa2

• added new interface IFreezable to tell a node to freeze itself, in order to allow BatchNormalization to honor CloneFunction (..., parameters="constant")

  frankseide

  committed
  0270010

• Converted AggregationOp to pure function template

  thilow

  committed
  fd95477

• Changed "NeutralValue" to function templates

  thilow

  committed
  bd776dc

• using function overloading for neutral

  thilow

  committed
  80fdb8f

Pagination

PreviousNext