https://github.com/Microsoft/CNTK

sort by:
Revision Author Date Message Commit Date
3296d0b bug fix: overflow check for very large minibatches did not typecast correctly. 10 February 2018, 02:15:10 UTC
8326dfa bug fix: relPosition should be relative to epochSize 10 February 2018, 01:49:02 UTC
ffc34c2 new option --fromLatest, to allow automatic restart from checkpoint in Philly retries 10 February 2018, 01:20:45 UTC
489aa6f checkpointing now saves a latest tag 10 February 2018, 01:07:45 UTC
6537f78 changed checkpointing to fractions of epochs 10 February 2018, 00:50:52 UTC
dc6bfaa added code to reader API to report current sample position; towards an epochSize-based scheduling, checkpointing, etc. (removed a debug-leftover sleep()) 09 February 2018, 23:42:06 UTC
6ed5349 added a comment 09 February 2018, 22:07:59 UTC
c78abf7 Merge branch 'fseide/dynamite' of https://github.com/Microsoft/cntk into fseide/dynamite 09 February 2018, 21:48:35 UTC
d681d42 added an overflow check to Unpack() for the cast of time index into a float32 (which can overflow for large minibatches) 09 February 2018, 21:46:41 UTC
70c5eb6 changed pathnames to have and interpolate with environment variables, for running on Philly 09 February 2018, 21:44:01 UTC
442c033 (comment) 09 February 2018, 21:42:44 UTC
c0383c4 uncommented some checks 09 February 2018, 21:42:28 UTC
bfec707 cleaned up checking code in AutoBatch 09 February 2018, 21:40:23 UTC
c609923 bug fix: Makefile lib order should pick up explicit boost lib before generic system path 08 February 2018, 17:58:25 UTC
adfc1ea increased bucketingFactor and loss-reporting smoothing time constant; turned dumping model parameters into a new command dump_model (and for that, split out the Marian calls into separate functions) 07 February 2018, 05:32:15 UTC
f03a669 (added a little Python script to convert Dynamite parameter dumps to a Marian model readable by marian-decoder) 03 February 2018, 07:20:34 UTC
50a5d41 bug fix: PlainTextDeserializer::PlainTextSequenceData should hold a ref count to each chunk's data-index array too, to avoid an incorrect early free; renamed a few variables in SetMatrixFromCSCFormat() in prep of reusing it for CSR 02 February 2018, 06:07:15 UTC
66fd02f MT.cpp now prints call stack in case of crash 02 February 2018, 01:19:47 UTC
4155605 cleaned up SetMatrixFromCSCFormat(), it should now be the same as SetMatrixFromCSRFormat() 02 February 2018, 00:53:35 UTC
a7decff cleaned up resetting of GPUSparseMatrix::m_sliceViewOffset 02 February 2018, 00:19:20 UTC
bfe74b4 CopyToCPUSparseMatrix() now makes a compact copy of the actual size, not preserving the allocation size 01 February 2018, 23:49:00 UTC
0972e98 bug fix in GatherBatch(): should use correct NzCount 01 February 2018, 22:04:51 UTC
1eec80c Matrix::AssignValuesOf() can now copy from GPUSparse to CPUSparse; bug fix: CopyToCPUSparseMatrix() should honor and adjust copied indices for sliceViewOffset 01 February 2018, 19:08:23 UTC
1931f11 added CheckSparse(), but not working 01 February 2018, 17:40:37 UTC
336b25f (comments) 01 February 2018, 08:41:49 UTC
365218c added lots of checks to NDArrayViewArena (seems all correctly working); disabled arena for now, to track down some bug 01 February 2018, 08:11:14 UTC
9ff0bb5 new methods MatrixBase::AsPtr(), AsRef(), ElemTypeIs() 01 February 2018, 08:08:39 UTC
6e6290d cleaned up the last commit 01 February 2018, 07:35:53 UTC
f6561c0 last commit improved by now launching more CUDA threads (more parallelism) 01 February 2018, 07:12:19 UTC
6885e3f refactored sparse Convolution with transposed rhs to move the loop into the CUDA kernel, avoiding lots of CUDA launches 01 February 2018, 06:58:04 UTC
5a95f30 Merge branch 'fseide/dynamite' of https://github.com/Microsoft/CNTK into fseide/dynamite 01 February 2018, 06:42:06 UTC
14451bd (comment) 01 February 2018, 06:42:03 UTC
aceeac6 clarified some names in sparse Convolution 01 February 2018, 06:41:08 UTC
3137490 renamed some variables in sparse kernels for clarity 01 February 2018, 06:39:42 UTC
4036e68 cleaned up sparse x dense -> dense, and added a Dynamite test 01 February 2018, 05:31:32 UTC
67bbbd5 removed s_recycledDenseArenas, as its role has been taken over by the recycled memory blocks 31 January 2018, 19:49:42 UTC
a9d4d30 removed s_currentArenaSize/Used 31 January 2018, 19:39:57 UTC
e447f57 removed s_currentArena 31 January 2018, 19:35:19 UTC
782e41a changed an indentation (separate commit since diff shows it as one huge changed blob), no code change 31 January 2018, 19:31:31 UTC
9d66ffb minor cleanup of previous commit, for better diffing 31 January 2018, 19:30:27 UTC
6315819 preparing to delete the arena allocation, moving to using gaps only; factored gap allocation into separate function (no code change, but some renaming) 31 January 2018, 19:25:07 UTC
f15bf2f (moved another function, no code change) 31 January 2018, 18:54:41 UTC
ffe27b5 (moved a function, no code change) 31 January 2018, 18:53:42 UTC
a0e8712 (minor fix to last commit) 31 January 2018, 18:51:54 UTC
a10c637 separated s_recycledArenass into one dense and sparse 31 January 2018, 18:49:18 UTC
d8a755b minor cleanup of last commit (in two steps for better diffing); bug fix: missed a lock_guard in previous refactoring 31 January 2018, 18:43:02 UTC
9d2d85c refactored NDArrayViewArena::New() a little, no code change otherwise 31 January 2018, 18:39:24 UTC
0d230c6 can now pass an allocator to Primitivefunction::Forward() and BackwardTo(), to allow temporary objects in the future; NewNDArrayView() now simply called New() 31 January 2018, 18:32:39 UTC
2930c48 bug fix: IsMatrixOfDataType() should use correct type 'double'; bug fix: GetSubBatches_CreateMinibatches() should account for the last sentence of a partial minibatch correctly; bug fix: Train() should not run minibatch source in inference mode! 31 January 2018, 17:59:43 UTC
66cfd80 rewrote partial batching, to account for Marian's padding 29 January 2018, 23:51:55 UTC
50dbc38 now catches bad_alloc and logs the specific MB 29 January 2018, 21:22:48 UTC
b4aa61b Merge branch 'fseide/dynamite' of https://github.com/Microsoft/cntk into fseide/dynamite 29 January 2018, 21:11:39 UTC
4934c9e TracingGPUMemoryAllocator::AllocateNoTrace(0 now throws bad_alloc(); simplified the allocator a little (most variables only exist for dense, so no need for arrays); bug fix: sparse allocation should not kill the current dense arena 29 January 2018, 21:11:30 UTC
499c3bc improved logging for mem alloc 29 January 2018, 19:45:36 UTC
cbbd830 merging now happens inside RecycleMemoryBlock() 29 January 2018, 19:12:32 UTC
9aaa78a refactored NewNDArray() quite a bit; bug fix: gap merging should increment j correctly in all cases 29 January 2018, 18:40:48 UTC
34c20fe added tracking to mem allocator; changed saveEvery 29 January 2018, 16:14:39 UTC
dc4756f super hacks in NewNDArrayView, seems to work, but could be sped up 28 January 2018, 20:46:45 UTC
dbf882d added diagnostics for allocator 28 January 2018, 02:55:14 UTC
17ee59d towards a gap allocator 28 January 2018, 02:51:45 UTC
b4d1930 prototypical implementation of recording recycled gaps 28 January 2018, 01:36:30 UTC
a01f79a disabled logging of gaps 28 January 2018, 01:11:08 UTC
cd183ed made gcc happy 28 January 2018, 00:43:00 UTC
adbaf0f added tracing to the memory allocator, in prep for more proper reuse of memory 28 January 2018, 00:38:16 UTC
ceaacce added a custom deleter to Matrix for matrix storage objects that wrap an external buffer. Not tested yet, but want to break this into small commits 27 January 2018, 18:47:26 UTC
5c4f1e2 (fixed a few messages); layer_norm() switched to Pow() 27 January 2018, 02:36:48 UTC
b5f8124 switched layer_norm to NormalizeDenormalize() 26 January 2018, 22:44:11 UTC
25da613 implemented NormalizeDenormalize() and tests 26 January 2018, 22:41:40 UTC
f2cd275 Merge branch 'fseide/dynamite' of https://github.com/Microsoft/CNTK into fseide/dynamite 26 January 2018, 21:57:06 UTC
853baa5 made gcc happy 26 January 2018, 21:56:53 UTC
163d024 removed some old, wrong version of matrix-weight backprop 26 January 2018, 21:53:37 UTC
f5625b6 added another test for ElementAffine() with an explicit formula, to test opAxBplusC itself 26 January 2018, 21:51:32 UTC
55226b4 implemented ElementAffine(), with tests; using it in layer_norm 26 January 2018, 21:49:03 UTC
348e73a Expr operators with scalars now use ScaleAndShift(), avoiding the constants 26 January 2018, 20:09:58 UTC
97607ad implemented ScaleAndShift(), with tests 26 January 2018, 19:54:52 UTC
e7c27aa bug fix: batched backprop into matrix weight can now merge more complex map shapes like found in Marian, by reshaping; changed marian::affine() to use Affine(), saving some memory 26 January 2018, 19:14:20 UTC
6c1085c (adapted a test to a new function signature) 26 January 2018, 19:10:23 UTC
c89cb67 undid an accidental commit of an intentional test failure 26 January 2018, 02:00:42 UTC
ed85c77 tests for variants of Affine() 26 January 2018, 01:57:47 UTC
be3d3e9 bug fix: gradient of TransposeAffine should be the same as TransposeTimes; added tests for variants of Affine() 26 January 2018, 01:53:15 UTC
9efe860 minor fix in marian::layer_norm 26 January 2018, 00:47:44 UTC
6d44662 implemented Affine() and a test. This required quite a bit of additions at several places; bug fix: back prop into matrix weight should use a || instead of && to match (this fix revealed another bug to be fixed later) 26 January 2018, 00:46:34 UTC
85f619f added stubs for new operations such as Affine() and ScaleAndShift(), but not implemented yet 25 January 2018, 22:27:07 UTC
6c9b6f9 added new PrimitiveFunction() constructor overload that takes an InputsVectorType 25 January 2018, 22:10:26 UTC
081aa04 added runtime stats to BackpropThroughSplice() and BackpropToMatrixWeight() 25 January 2018, 07:08:45 UTC
d8951a1 disabled LR decay; disabled reading out partial-MB loss from GPU 25 January 2018, 06:43:27 UTC
2c513db bug fixes: when passing something by move(), the same argument list should not access it; added statistics reporting for BackpropToUnbatched() 25 January 2018, 05:50:39 UTC
e917e77 towards moving backprop onto bg thread 25 January 2018, 03:30:43 UTC
cc7c591 cleaned up error handling in the Memoizer thread 25 January 2018, 03:17:25 UTC
8a1f8e1 minor updates of logging 25 January 2018, 01:42:42 UTC
f20678a fix in AutoBatch stats-logging counter 25 January 2018, 01:35:20 UTC
a2d588a BatchedBackward() now collects init time; control of batching stats fixed w.r.t. backward 25 January 2018, 01:21:32 UTC
85d737f (fixed statistics) 25 January 2018, 01:11:25 UTC
163353f towards showsing stats for Backward 24 January 2018, 19:54:24 UTC
9c88fba disabled GPU sync for time measurement in BatcgedForward(); removed the hard-coded Marian hyper-parameters in Adam; MT: changed savedEvery to 10000; updated some logging; changed Adam momentum parameters to better match Marian's; added a new end-to-end timer for the non-Update() sub-minibatches 24 January 2018, 19:39:14 UTC
e780fd3 Merge branch 'fseide/dynamite' of https://github.com/Microsoft/CNTK into fseide/dynamite 24 January 2018, 16:57:30 UTC
6c4a2f2 gcc compiler happiness 24 January 2018, 16:57:23 UTC
00c5592 bug fix: RAggregateGradientFromAllConsumers() should separate transposed and non-transposed Times for weight gradient, to ensure the dense gradients are done before the sparse ones for the same weight; bug fix: marian DropoutMask() should scale the mask; reenabled dropout in Transformer 24 January 2018, 16:51:09 UTC
d7a36ba adjusted the tests for BernoulliRandom() 23 January 2018, 22:10:11 UTC
80fa470 Marian dropout now implemented using BernoulliRandom() 23 January 2018, 09:48:06 UTC
back to top