CPU info:
    CPU Model Name: Intel(R) Xeon(R) CPU E5-2690 v3 @ 2.60GHz
    Hardware threads: 12
    Total Memory: 57700428 kB
-------------------------------------------------------------------
=== Running mpiexec -n 2 /home/ubuntu/workspace/build/gpu/release/bin/cntk configFile=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/../ParallelBM/cntk.cntk currentDirectory=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data RunDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu DataDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data ConfigDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/.. OutputDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu DeviceId=-1 timestamping=true numCPUThreads=6 precision=double speechTrain=[SGD=[ParallelTrain=[parallelizationStartEpoch=2]]] stderr=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/stderr
CNTK 2.3.1+ (HEAD f4f0f8, Dec 11 2017 18:34:12) at 2017/12/12 15:17:58

/home/ubuntu/workspace/build/gpu/release/bin/cntk  configFile=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/../ParallelBM/cntk.cntk  currentDirectory=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data  RunDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu  DataDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data  ConfigDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/..  OutputDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu  DeviceId=-1  timestamping=true  numCPUThreads=6  precision=double  speechTrain=[SGD=[ParallelTrain=[parallelizationStartEpoch=2]]]  stderr=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/stderr
CNTK 2.3.1+ (HEAD f4f0f8, Dec 11 2017 18:34:12) at 2017/12/12 15:17:58

/home/ubuntu/workspace/build/gpu/release/bin/cntk  configFile=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/../ParallelBM/cntk.cntk  currentDirectory=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data  RunDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu  DataDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data  ConfigDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/..  OutputDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu  DeviceId=-1  timestamping=true  numCPUThreads=6  precision=double  speechTrain=[SGD=[ParallelTrain=[parallelizationStartEpoch=2]]]  stderr=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/stderr
Changed current directory to /home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data
Changed current directory to /home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data
--------------------------------------------------------------------------
[[36520,1],1]: A high-performance Open MPI point-to-point messaging module
was unable to find any relevant network interfaces:

Module: OpenFabrics (openib)
  Host: fdb4dbbde386

Another transport will be used instead, although this may result in
lower performance.
--------------------------------------------------------------------------
ping [requestnodes (before change)]: 2 nodes pinging each other
ping [requestnodes (before change)]: 2 nodes pinging each other
ping [requestnodes (after change)]: 2 nodes pinging each other
requestnodes [MPIWrapperMpi]: using 2 out of 2 MPI nodes on a single host (2 requested); we (0) are in (participating)
ping [mpihelper]: 2 nodes pinging each other
ping [requestnodes (after change)]: 2 nodes pinging each other
requestnodes [MPIWrapperMpi]: using 2 out of 2 MPI nodes on a single host (2 requested); we (1) are in (participating)
ping [mpihelper]: 2 nodes pinging each other
12/12/2017 15:17:58: Redirecting stderr to file /tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/stderr_speechTrain.logrank0
12/12/2017 15:17:58: Redirecting stderr to file /tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/stderr_speechTrain.logrank1
[fdb4dbbde386:77975] 1 more process has sent help message help-mpi-btl-base.txt / btl:no-nics
[fdb4dbbde386:77975] Set MCA parameter "orte_base_help_aggregate" to 0 to see all help / error messages
MPI Rank 0: CNTK 2.3.1+ (HEAD f4f0f8, Dec 11 2017 18:34:12) at 2017/12/12 15:17:58
MPI Rank 0: 
MPI Rank 0: /home/ubuntu/workspace/build/gpu/release/bin/cntk  configFile=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/../ParallelBM/cntk.cntk  currentDirectory=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data  RunDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu  DataDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data  ConfigDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/..  OutputDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu  DeviceId=-1  timestamping=true  numCPUThreads=6  precision=double  speechTrain=[SGD=[ParallelTrain=[parallelizationStartEpoch=2]]]  stderr=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/stderr
MPI Rank 0: 12/12/2017 15:17:58: -------------------------------------------------------------------
MPI Rank 0: 12/12/2017 15:17:58: Build info: 
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:17:58: 		Built time: Dec 11 2017 18:28:39
MPI Rank 0: 12/12/2017 15:17:58: 		Last modified date: Wed Nov 15 09:27:10 2017
MPI Rank 0: 12/12/2017 15:17:58: 		Build type: release
MPI Rank 0: 12/12/2017 15:17:58: 		Build target: GPU
MPI Rank 0: 12/12/2017 15:17:58: 		With ASGD: yes
MPI Rank 0: 12/12/2017 15:17:58: 		Math lib: mkl
MPI Rank 0: 12/12/2017 15:17:58: 		CUDA version: 9.0.0
MPI Rank 0: 12/12/2017 15:17:58: 		CUDNN version: 7.0.4
MPI Rank 0: 12/12/2017 15:17:58: 		Build Branch: HEAD
MPI Rank 0: 12/12/2017 15:17:58: 		Build SHA1: f4f0f82eabcc482dbd03af1f946a44ae2b8b97bf
MPI Rank 0: 12/12/2017 15:17:58: 		MPI distribution: Open MPI
MPI Rank 0: 12/12/2017 15:17:58: 		MPI version: 1.10.7
MPI Rank 0: 12/12/2017 15:17:58: -------------------------------------------------------------------
MPI Rank 0: 12/12/2017 15:17:58: -------------------------------------------------------------------
MPI Rank 0: 12/12/2017 15:17:58: GPU info:
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:17:58: 		Device[0]: cores = 3072; computeCapability = 5.2; type = "Tesla M60"; total memory = 8123 MB; free memory = 8112 MB
MPI Rank 0: 12/12/2017 15:17:58: -------------------------------------------------------------------
MPI Rank 0: 12/12/2017 15:17:58: Using 6 CPU threads.
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:17:58: ##############################################################################
MPI Rank 0: 12/12/2017 15:17:58: #                                                                            #
MPI Rank 0: 12/12/2017 15:17:58: # speechTrain command (train action)                                         #
MPI Rank 0: 12/12/2017 15:17:58: #                                                                            #
MPI Rank 0: 12/12/2017 15:17:58: ##############################################################################
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:17:58: 
MPI Rank 0: Creating virgin network.
MPI Rank 0: SimpleNetworkBuilder Using CPU
MPI Rank 0: reading script file glob_0000.scp ... 948 entries
MPI Rank 0: total 132 state names in state list /home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data/state.list
MPI Rank 0: htkmlfreader: reading MLF file /home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data/glob_0000.mlf ... total 948 entries
MPI Rank 0: ...............................................................................................feature set 0: 252734 frames in 948 out of 948 utterances
MPI Rank 0: label set 0: 129 classes
MPI Rank 0: minibatchutterancesource: 948 utterances grouped into 3 chunks, av. chunk size: 316.0 utterances, 84244.7 frames
MPI Rank 0: 12/12/2017 15:17:58: 
MPI Rank 0: Model has 25 nodes. Using CPU.
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:17:58: Training criterion:   CrossEntropyWithSoftmax = CrossEntropyWithSoftmax
MPI Rank 0: 12/12/2017 15:17:58: Evaluation criterion: EvalClassificationError = ClassificationError
MPI Rank 0: 
MPI Rank 0: 
MPI Rank 0: Allocating matrices for forward and/or backward propagation.
MPI Rank 0: 
MPI Rank 0: Gradient Memory Aliasing: 4 are aliased.
MPI Rank 0: 	W2*H1 (gradient) reuses HLast (gradient)
MPI Rank 0: 	W1*H1 (gradient) reuses W1*H1+B1 (gradient)
MPI Rank 0: 
MPI Rank 0: Memory Sharing: Out of 40 matrices, 21 are shared as 5, and 19 are not shared.
MPI Rank 0: 
MPI Rank 0: Here are the ones that share memory:
MPI Rank 0: 	{ PosteriorProb : [132 x 1 x *]
MPI Rank 0: 	  ScaledLogLikelihood : [132 x 1 x *] }
MPI Rank 0: 	{ H2 : [512 x 1 x *]
MPI Rank 0: 	  W0*features+B0 : [512 x 1 x *]
MPI Rank 0: 	  W1 : [512 x 512] (gradient)
MPI Rank 0: 	  W1*H1 : [512 x 1 x *] }
MPI Rank 0: 	{ HLast : [132 x 1 x *] (gradient)
MPI Rank 0: 	  W0 : [512 x 363] (gradient)
MPI Rank 0: 	  W0*features+B0 : [512 x 1 x *] (gradient)
MPI Rank 0: 	  W1*H1 : [512 x 1 x *] (gradient)
MPI Rank 0: 	  W1*H1+B1 : [512 x 1 x *]
MPI Rank 0: 	  W1*H1+B1 : [512 x 1 x *] (gradient)
MPI Rank 0: 	  W2*H1 : [132 x 1 x *]
MPI Rank 0: 	  W2*H1 : [132 x 1 x *] (gradient) }
MPI Rank 0: 	{ H1 : [512 x 1 x *] (gradient)
MPI Rank 0: 	  H2 : [512 x 1 x *] (gradient)
MPI Rank 0: 	  HLast : [132 x 1 x *]
MPI Rank 0: 	  W0*features : [512 x *]
MPI Rank 0: 	  W0*features : [512 x *] (gradient) }
MPI Rank 0: 	{ B0 : [512 x 1] (gradient)
MPI Rank 0: 	  H1 : [512 x 1 x *] }
MPI Rank 0: 
MPI Rank 0: Here are the ones that don't share memory:
MPI Rank 0: 	{MeanOfFeatures : [363]}
MPI Rank 0: 	{InvStdOfFeatures : [363]}
MPI Rank 0: 	{features : [363 x *]}
MPI Rank 0: 	{W0 : [512 x 363]}
MPI Rank 0: 	{B0 : [512 x 1]}
MPI Rank 0: 	{W1 : [512 x 512]}
MPI Rank 0: 	{B1 : [512 x 1]}
MPI Rank 0: 	{W2 : [132 x 512]}
MPI Rank 0: 	{B2 : [132 x 1]}
MPI Rank 0: 	{labels : [132 x *]}
MPI Rank 0: 	{Prior : [132]}
MPI Rank 0: 	{CrossEntropyWithSoftmax : [1]}
MPI Rank 0: 	{B1 : [512 x 1] (gradient)}
MPI Rank 0: 	{CrossEntropyWithSoftmax : [1] (gradient)}
MPI Rank 0: 	{LogOfPrior : [132]}
MPI Rank 0: 	{B2 : [132 x 1] (gradient)}
MPI Rank 0: 	{W2 : [132 x 512] (gradient)}
MPI Rank 0: 	{EvalClassificationError : [1]}
MPI Rank 0: 	{MVNormalizedFeatures : [363 x *]}
MPI Rank 0: 
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:17:58: Training 516740 parameters in 6 out of 6 parameter tensors and 15 nodes with gradient:
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:17:58: 	Node 'B0' (LearnableParameter operation) : [512 x 1]
MPI Rank 0: 12/12/2017 15:17:58: 	Node 'B1' (LearnableParameter operation) : [512 x 1]
MPI Rank 0: 12/12/2017 15:17:58: 	Node 'B2' (LearnableParameter operation) : [132 x 1]
MPI Rank 0: 12/12/2017 15:17:58: 	Node 'W0' (LearnableParameter operation) : [512 x 363]
MPI Rank 0: 12/12/2017 15:17:58: 	Node 'W1' (LearnableParameter operation) : [512 x 512]
MPI Rank 0: 12/12/2017 15:17:58: 	Node 'W2' (LearnableParameter operation) : [132 x 512]
MPI Rank 0: 
MPI Rank 0: NcclComm: disabled, at least one rank using CPU device
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:17:59: Precomputing --> 3 PreCompute nodes found.
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:17:59: 	MeanOfFeatures = Mean()
MPI Rank 0: 12/12/2017 15:17:59: 	InvStdOfFeatures = InvStdDev()
MPI Rank 0: 12/12/2017 15:17:59: 	Prior = Mean()
MPI Rank 0: minibatchiterator: epoch 0: frames [0..252734] (first utterance at frame 0), data subset 0 of 1, with 1 datapasses
MPI Rank 0: requiredata: determined feature kind as 33-dimensional 'USER' with frame shift 10.0 ms
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:04: Precomputing --> Completed.
MPI Rank 0: 
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:04: Starting Epoch 1: learning rate per sample = 0.015625  effective momentum = 0.900000  momentum as time constant = 607.4 samples
MPI Rank 0: minibatchiterator: epoch 0: frames [0..20480] (first utterance at frame 0), data subset 0 of 1, with 1 datapasses
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:04: Starting minibatch loop.
MPI Rank 0: 12/12/2017 15:18:04:  Epoch[ 1 of 5]-Minibatch[   1-   3, 0.94%]: CrossEntropyWithSoftmax = 4.65054456 * 192; EvalClassificationError = 0.96875000 * 192; time = 0.1066s; samplesPerSecond = 1801.3
MPI Rank 0: 12/12/2017 15:18:04:  Epoch[ 1 of 5]-Minibatch[   4-   6, 1.88%]: CrossEntropyWithSoftmax = 4.41918514 * 192; EvalClassificationError = 0.88020833 * 192; time = 0.0582s; samplesPerSecond = 3297.3
MPI Rank 0: 12/12/2017 15:18:04:  Epoch[ 1 of 5]-Minibatch[   7-   9, 2.81%]: CrossEntropyWithSoftmax = 4.76293439 * 192; EvalClassificationError = 0.93229167 * 192; time = 0.0572s; samplesPerSecond = 3354.1
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  10-  12, 3.75%]: CrossEntropyWithSoftmax = 4.23168138 * 192; EvalClassificationError = 0.90104167 * 192; time = 0.0558s; samplesPerSecond = 3441.3
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  13-  15, 4.69%]: CrossEntropyWithSoftmax = 4.47386985 * 192; EvalClassificationError = 0.89583333 * 192; time = 0.0588s; samplesPerSecond = 3264.4
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  16-  18, 5.62%]: CrossEntropyWithSoftmax = 4.39795080 * 192; EvalClassificationError = 0.94270833 * 192; time = 0.0555s; samplesPerSecond = 3460.5
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  19-  21, 6.56%]: CrossEntropyWithSoftmax = 4.25148851 * 192; EvalClassificationError = 0.97395833 * 192; time = 0.0558s; samplesPerSecond = 3438.8
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  22-  24, 7.50%]: CrossEntropyWithSoftmax = 4.06906673 * 192; EvalClassificationError = 0.89062500 * 192; time = 0.2700s; samplesPerSecond = 711.2
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  25-  27, 8.44%]: CrossEntropyWithSoftmax = 3.98165780 * 192; EvalClassificationError = 0.90104167 * 192; time = 0.0563s; samplesPerSecond = 3412.2
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  28-  30, 9.38%]: CrossEntropyWithSoftmax = 3.84788960 * 192; EvalClassificationError = 0.85416667 * 192; time = 0.0565s; samplesPerSecond = 3397.8
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  31-  33, 10.31%]: CrossEntropyWithSoftmax = 3.80540005 * 192; EvalClassificationError = 0.84895833 * 192; time = 0.0553s; samplesPerSecond = 3474.6
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  34-  36, 11.25%]: CrossEntropyWithSoftmax = 3.76164351 * 192; EvalClassificationError = 0.86979167 * 192; time = 0.0540s; samplesPerSecond = 3555.9
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  37-  39, 12.19%]: CrossEntropyWithSoftmax = 3.74042928 * 192; EvalClassificationError = 0.82812500 * 192; time = 0.0553s; samplesPerSecond = 3474.8
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  40-  42, 13.12%]: CrossEntropyWithSoftmax = 3.69572235 * 192; EvalClassificationError = 0.83333333 * 192; time = 0.0557s; samplesPerSecond = 3449.1
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  43-  45, 14.06%]: CrossEntropyWithSoftmax = 3.89722237 * 192; EvalClassificationError = 0.90104167 * 192; time = 0.0552s; samplesPerSecond = 3481.0
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  46-  48, 15.00%]: CrossEntropyWithSoftmax = 3.77423768 * 192; EvalClassificationError = 0.84375000 * 192; time = 0.0539s; samplesPerSecond = 3563.5
MPI Rank 0: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  49-  51, 15.94%]: CrossEntropyWithSoftmax = 3.81262710 * 192; EvalClassificationError = 0.88541667 * 192; time = 0.0563s; samplesPerSecond = 3411.4
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  52-  54, 16.88%]: CrossEntropyWithSoftmax = 3.92558806 * 192; EvalClassificationError = 0.88020833 * 192; time = 0.0545s; samplesPerSecond = 3520.9
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  55-  57, 17.81%]: CrossEntropyWithSoftmax = 3.56995707 * 192; EvalClassificationError = 0.84895833 * 192; time = 0.0580s; samplesPerSecond = 3312.2
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  58-  60, 18.75%]: CrossEntropyWithSoftmax = 3.67092670 * 192; EvalClassificationError = 0.88541667 * 192; time = 0.0552s; samplesPerSecond = 3476.5
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  61-  63, 19.69%]: CrossEntropyWithSoftmax = 3.36174584 * 192; EvalClassificationError = 0.77083333 * 192; time = 0.0568s; samplesPerSecond = 3380.7
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  64-  66, 20.62%]: CrossEntropyWithSoftmax = 3.42123462 * 192; EvalClassificationError = 0.81770833 * 192; time = 0.0551s; samplesPerSecond = 3482.9
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  67-  69, 21.56%]: CrossEntropyWithSoftmax = 3.36684019 * 192; EvalClassificationError = 0.79166667 * 192; time = 0.0559s; samplesPerSecond = 3432.4
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  70-  72, 22.50%]: CrossEntropyWithSoftmax = 3.68558416 * 192; EvalClassificationError = 0.84375000 * 192; time = 0.0574s; samplesPerSecond = 3343.8
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  73-  75, 23.44%]: CrossEntropyWithSoftmax = 3.64969026 * 192; EvalClassificationError = 0.82812500 * 192; time = 0.0552s; samplesPerSecond = 3475.5
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  76-  78, 24.38%]: CrossEntropyWithSoftmax = 3.37812633 * 192; EvalClassificationError = 0.81770833 * 192; time = 0.0566s; samplesPerSecond = 3395.2
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  79-  81, 25.31%]: CrossEntropyWithSoftmax = 3.52781334 * 192; EvalClassificationError = 0.82291667 * 192; time = 0.0552s; samplesPerSecond = 3479.4
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  82-  84, 26.25%]: CrossEntropyWithSoftmax = 3.47704383 * 192; EvalClassificationError = 0.81250000 * 192; time = 0.0577s; samplesPerSecond = 3325.9
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  85-  87, 27.19%]: CrossEntropyWithSoftmax = 3.56621904 * 192; EvalClassificationError = 0.82812500 * 192; time = 0.0590s; samplesPerSecond = 3253.4
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  88-  90, 28.12%]: CrossEntropyWithSoftmax = 3.44116285 * 192; EvalClassificationError = 0.80729167 * 192; time = 0.0553s; samplesPerSecond = 3471.8
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  91-  93, 29.06%]: CrossEntropyWithSoftmax = 3.37856471 * 192; EvalClassificationError = 0.80729167 * 192; time = 0.0557s; samplesPerSecond = 3445.6
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  94-  96, 30.00%]: CrossEntropyWithSoftmax = 3.56985531 * 192; EvalClassificationError = 0.82291667 * 192; time = 0.0570s; samplesPerSecond = 3367.7
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  97-  99, 30.94%]: CrossEntropyWithSoftmax = 3.36523472 * 192; EvalClassificationError = 0.81770833 * 192; time = 0.0557s; samplesPerSecond = 3447.4
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[ 100- 102, 31.87%]: CrossEntropyWithSoftmax = 3.51274161 * 192; EvalClassificationError = 0.82291667 * 192; time = 0.0558s; samplesPerSecond = 3443.5
MPI Rank 0: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[ 103- 105, 32.81%]: CrossEntropyWithSoftmax = 3.57161359 * 192; EvalClassificationError = 0.83333333 * 192; time = 0.0575s; samplesPerSecond = 3339.9
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 106- 108, 33.75%]: CrossEntropyWithSoftmax = 3.50343297 * 192; EvalClassificationError = 0.82812500 * 192; time = 0.0548s; samplesPerSecond = 3506.8
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 109- 111, 34.69%]: CrossEntropyWithSoftmax = 3.34024400 * 192; EvalClassificationError = 0.79687500 * 192; time = 0.0579s; samplesPerSecond = 3318.1
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 112- 114, 35.62%]: CrossEntropyWithSoftmax = 3.30238305 * 192; EvalClassificationError = 0.80729167 * 192; time = 0.0545s; samplesPerSecond = 3522.3
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 115- 117, 36.56%]: CrossEntropyWithSoftmax = 3.21012106 * 192; EvalClassificationError = 0.78645833 * 192; time = 0.0610s; samplesPerSecond = 3149.8
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 118- 120, 37.50%]: CrossEntropyWithSoftmax = 3.27497388 * 192; EvalClassificationError = 0.77604167 * 192; time = 0.0559s; samplesPerSecond = 3435.6
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 121- 123, 38.44%]: CrossEntropyWithSoftmax = 3.55026166 * 192; EvalClassificationError = 0.83854167 * 192; time = 0.0550s; samplesPerSecond = 3492.9
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 124- 126, 39.38%]: CrossEntropyWithSoftmax = 3.01866034 * 192; EvalClassificationError = 0.73958333 * 192; time = 0.2717s; samplesPerSecond = 706.6
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 127- 129, 40.31%]: CrossEntropyWithSoftmax = 3.13818688 * 192; EvalClassificationError = 0.80208333 * 192; time = 0.0543s; samplesPerSecond = 3537.7
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 130- 132, 41.25%]: CrossEntropyWithSoftmax = 3.07582410 * 192; EvalClassificationError = 0.76041667 * 192; time = 0.0531s; samplesPerSecond = 3614.3
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 133- 135, 42.19%]: CrossEntropyWithSoftmax = 3.19786112 * 192; EvalClassificationError = 0.75520833 * 192; time = 0.0597s; samplesPerSecond = 3218.1
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 136- 138, 43.12%]: CrossEntropyWithSoftmax = 2.79776977 * 192; EvalClassificationError = 0.71875000 * 192; time = 0.0592s; samplesPerSecond = 3242.3
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 139- 141, 44.06%]: CrossEntropyWithSoftmax = 3.19303582 * 192; EvalClassificationError = 0.82291667 * 192; time = 0.0658s; samplesPerSecond = 2918.0
MPI Rank 0: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 142- 144, 45.00%]: CrossEntropyWithSoftmax = 3.04555903 * 192; EvalClassificationError = 0.75520833 * 192; time = 0.0686s; samplesPerSecond = 2800.1
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 145- 147, 45.94%]: CrossEntropyWithSoftmax = 3.03596679 * 192; EvalClassificationError = 0.72916667 * 192; time = 0.0608s; samplesPerSecond = 3156.0
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 148- 150, 46.88%]: CrossEntropyWithSoftmax = 2.81093303 * 192; EvalClassificationError = 0.64583333 * 192; time = 0.0571s; samplesPerSecond = 3363.6
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 151- 153, 47.81%]: CrossEntropyWithSoftmax = 3.05364288 * 192; EvalClassificationError = 0.69791667 * 192; time = 0.0571s; samplesPerSecond = 3364.8
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 154- 156, 48.75%]: CrossEntropyWithSoftmax = 3.18200713 * 192; EvalClassificationError = 0.78645833 * 192; time = 0.0578s; samplesPerSecond = 3319.0
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 157- 159, 49.69%]: CrossEntropyWithSoftmax = 3.13461023 * 192; EvalClassificationError = 0.75000000 * 192; time = 0.0555s; samplesPerSecond = 3461.2
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 160- 162, 50.62%]: CrossEntropyWithSoftmax = 2.88863478 * 192; EvalClassificationError = 0.72395833 * 192; time = 0.0571s; samplesPerSecond = 3362.9
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 163- 165, 51.56%]: CrossEntropyWithSoftmax = 2.81095072 * 192; EvalClassificationError = 0.70833333 * 192; time = 0.0572s; samplesPerSecond = 3354.2
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 166- 168, 52.50%]: CrossEntropyWithSoftmax = 3.02416042 * 192; EvalClassificationError = 0.73958333 * 192; time = 0.0583s; samplesPerSecond = 3290.7
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 169- 171, 53.44%]: CrossEntropyWithSoftmax = 2.75149996 * 192; EvalClassificationError = 0.69791667 * 192; time = 0.0554s; samplesPerSecond = 3465.8
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 172- 174, 54.37%]: CrossEntropyWithSoftmax = 2.66519087 * 192; EvalClassificationError = 0.65625000 * 192; time = 0.0531s; samplesPerSecond = 3613.5
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 175- 177, 55.31%]: CrossEntropyWithSoftmax = 2.73446036 * 192; EvalClassificationError = 0.65625000 * 192; time = 0.0567s; samplesPerSecond = 3389.0
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 178- 180, 56.25%]: CrossEntropyWithSoftmax = 2.76060156 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0558s; samplesPerSecond = 3439.8
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 181- 183, 57.19%]: CrossEntropyWithSoftmax = 2.84893104 * 192; EvalClassificationError = 0.73958333 * 192; time = 0.0583s; samplesPerSecond = 3295.0
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 184- 186, 58.13%]: CrossEntropyWithSoftmax = 2.82041431 * 192; EvalClassificationError = 0.72916667 * 192; time = 0.0585s; samplesPerSecond = 3282.8
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 187- 189, 59.06%]: CrossEntropyWithSoftmax = 2.81359166 * 192; EvalClassificationError = 0.72916667 * 192; time = 0.0553s; samplesPerSecond = 3471.3
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 190- 192, 60.00%]: CrossEntropyWithSoftmax = 2.80548960 * 192; EvalClassificationError = 0.71354167 * 192; time = 0.0574s; samplesPerSecond = 3343.8
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 193- 195, 60.94%]: CrossEntropyWithSoftmax = 2.57693072 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0566s; samplesPerSecond = 3390.7
MPI Rank 0: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 196- 198, 61.88%]: CrossEntropyWithSoftmax = 3.00545481 * 192; EvalClassificationError = 0.74479167 * 192; time = 0.0572s; samplesPerSecond = 3358.6
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 199- 201, 62.81%]: CrossEntropyWithSoftmax = 2.66074209 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0591s; samplesPerSecond = 3249.7
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 202- 204, 63.75%]: CrossEntropyWithSoftmax = 2.55576220 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0589s; samplesPerSecond = 3261.6
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 205- 207, 64.69%]: CrossEntropyWithSoftmax = 2.63325580 * 192; EvalClassificationError = 0.66666667 * 192; time = 0.0552s; samplesPerSecond = 3480.5
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 208- 210, 65.62%]: CrossEntropyWithSoftmax = 2.59688083 * 192; EvalClassificationError = 0.67187500 * 192; time = 0.0592s; samplesPerSecond = 3243.6
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 211- 213, 66.56%]: CrossEntropyWithSoftmax = 2.59744697 * 192; EvalClassificationError = 0.61979167 * 192; time = 0.0569s; samplesPerSecond = 3374.1
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 214- 216, 67.50%]: CrossEntropyWithSoftmax = 2.49590166 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0544s; samplesPerSecond = 3528.1
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 217- 219, 68.44%]: CrossEntropyWithSoftmax = 2.75606293 * 192; EvalClassificationError = 0.67708333 * 192; time = 0.0644s; samplesPerSecond = 2981.8
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 220- 222, 69.38%]: CrossEntropyWithSoftmax = 2.51043915 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0571s; samplesPerSecond = 3363.2
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 223- 225, 70.31%]: CrossEntropyWithSoftmax = 2.46191162 * 192; EvalClassificationError = 0.66145833 * 192; time = 0.0575s; samplesPerSecond = 3340.3
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 226- 228, 71.25%]: CrossEntropyWithSoftmax = 2.74930663 * 192; EvalClassificationError = 0.70312500 * 192; time = 0.2631s; samplesPerSecond = 729.9
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 229- 231, 72.19%]: CrossEntropyWithSoftmax = 2.56948343 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0556s; samplesPerSecond = 3455.1
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 232- 234, 73.12%]: CrossEntropyWithSoftmax = 2.99801669 * 192; EvalClassificationError = 0.77083333 * 192; time = 0.0561s; samplesPerSecond = 3422.6
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 235- 237, 74.06%]: CrossEntropyWithSoftmax = 2.37847319 * 192; EvalClassificationError = 0.59895833 * 192; time = 0.0567s; samplesPerSecond = 3383.4
MPI Rank 0: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 238- 240, 75.00%]: CrossEntropyWithSoftmax = 2.26592694 * 192; EvalClassificationError = 0.61979167 * 192; time = 0.0572s; samplesPerSecond = 3358.0
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 241- 243, 75.94%]: CrossEntropyWithSoftmax = 2.30186997 * 192; EvalClassificationError = 0.57291667 * 192; time = 0.0591s; samplesPerSecond = 3246.2
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 244- 246, 76.88%]: CrossEntropyWithSoftmax = 2.70240793 * 192; EvalClassificationError = 0.70312500 * 192; time = 0.0582s; samplesPerSecond = 3299.9
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 247- 249, 77.81%]: CrossEntropyWithSoftmax = 2.43935470 * 192; EvalClassificationError = 0.60937500 * 192; time = 0.0547s; samplesPerSecond = 3507.4
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 250- 252, 78.75%]: CrossEntropyWithSoftmax = 2.52037152 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0580s; samplesPerSecond = 3311.2
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 253- 255, 79.69%]: CrossEntropyWithSoftmax = 2.38274509 * 192; EvalClassificationError = 0.63020833 * 192; time = 0.0562s; samplesPerSecond = 3417.8
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 256- 258, 80.62%]: CrossEntropyWithSoftmax = 2.36861217 * 192; EvalClassificationError = 0.59375000 * 192; time = 0.0555s; samplesPerSecond = 3457.2
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 259- 261, 81.56%]: CrossEntropyWithSoftmax = 2.34453624 * 192; EvalClassificationError = 0.63541667 * 192; time = 0.0556s; samplesPerSecond = 3453.6
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 262- 264, 82.50%]: CrossEntropyWithSoftmax = 2.29446007 * 192; EvalClassificationError = 0.59895833 * 192; time = 0.0610s; samplesPerSecond = 3145.2
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 265- 267, 83.44%]: CrossEntropyWithSoftmax = 2.09108193 * 192; EvalClassificationError = 0.52604167 * 192; time = 0.0559s; samplesPerSecond = 3436.3
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 268- 270, 84.38%]: CrossEntropyWithSoftmax = 2.42635931 * 192; EvalClassificationError = 0.66145833 * 192; time = 0.0622s; samplesPerSecond = 3088.8
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 271- 273, 85.31%]: CrossEntropyWithSoftmax = 2.53475002 * 192; EvalClassificationError = 0.66145833 * 192; time = 0.0597s; samplesPerSecond = 3215.4
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 274- 276, 86.25%]: CrossEntropyWithSoftmax = 2.35308728 * 192; EvalClassificationError = 0.61458333 * 192; time = 0.0562s; samplesPerSecond = 3415.3
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 277- 279, 87.19%]: CrossEntropyWithSoftmax = 2.62700347 * 192; EvalClassificationError = 0.69270833 * 192; time = 0.0568s; samplesPerSecond = 3377.4
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 280- 282, 88.12%]: CrossEntropyWithSoftmax = 2.48479326 * 192; EvalClassificationError = 0.61458333 * 192; time = 0.0566s; samplesPerSecond = 3389.4
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 283- 285, 89.06%]: CrossEntropyWithSoftmax = 2.31729432 * 192; EvalClassificationError = 0.62500000 * 192; time = 0.0561s; samplesPerSecond = 3422.7
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 286- 288, 90.00%]: CrossEntropyWithSoftmax = 2.21677298 * 192; EvalClassificationError = 0.58854167 * 192; time = 0.0556s; samplesPerSecond = 3454.7
MPI Rank 0: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 289- 291, 90.94%]: CrossEntropyWithSoftmax = 2.28114207 * 192; EvalClassificationError = 0.58333333 * 192; time = 0.0578s; samplesPerSecond = 3321.7
MPI Rank 0: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 292- 294, 91.88%]: CrossEntropyWithSoftmax = 2.12055459 * 192; EvalClassificationError = 0.57812500 * 192; time = 0.0553s; samplesPerSecond = 3469.2
MPI Rank 0: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 295- 297, 92.81%]: CrossEntropyWithSoftmax = 2.31557649 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0588s; samplesPerSecond = 3263.0
MPI Rank 0: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 298- 300, 93.75%]: CrossEntropyWithSoftmax = 2.30559259 * 192; EvalClassificationError = 0.60937500 * 192; time = 0.0580s; samplesPerSecond = 3311.7
MPI Rank 0: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 301- 303, 94.69%]: CrossEntropyWithSoftmax = 2.15644939 * 192; EvalClassificationError = 0.58854167 * 192; time = 0.0590s; samplesPerSecond = 3253.1
MPI Rank 0: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 304- 306, 95.62%]: CrossEntropyWithSoftmax = 2.26862225 * 192; EvalClassificationError = 0.58333333 * 192; time = 0.0573s; samplesPerSecond = 3349.7
MPI Rank 0: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 307- 309, 96.56%]: CrossEntropyWithSoftmax = 2.15494679 * 192; EvalClassificationError = 0.54166667 * 192; time = 0.0567s; samplesPerSecond = 3389.0
MPI Rank 0: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 310- 312, 97.50%]: CrossEntropyWithSoftmax = 2.44588898 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0567s; samplesPerSecond = 3388.7
MPI Rank 0: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 313- 315, 98.44%]: CrossEntropyWithSoftmax = 2.17852500 * 192; EvalClassificationError = 0.58854167 * 192; time = 0.0597s; samplesPerSecond = 3215.4
MPI Rank 0: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 316- 318, 99.38%]: CrossEntropyWithSoftmax = 2.23007043 * 192; EvalClassificationError = 0.56770833 * 192; time = 0.0546s; samplesPerSecond = 3518.2
MPI Rank 0: 12/12/2017 15:18:11: Finished Epoch[ 1 of 5]: [Training] CrossEntropyWithSoftmax = 3.04696987 * 20480; EvalClassificationError = 0.73583984 * 20480; totalSamplesSeen = 20480; learningRatePerSample = 0.015625; epochTime=6.77772s
MPI Rank 0: 12/12/2017 15:18:11: SGD: Saving checkpoint model '/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/models/cntkSpeech.dnn.1'
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:11: Starting Epoch 2: learning rate per sample = 0.001953  effective momentum = 0.656119  momentum as time constant = 607.5 samples
MPI Rank 0: Parallel training (2 workers) using BlockMomentumSGD with block momentum = 0.5000, block momentum time constant (per worker) = 2954.6394, block learning rate = 1.0000, block size per worker = 2048 samples, using Nesterov-style block momentum, resetting SGD momentum after sync.
MPI Rank 0: minibatchiterator: epoch 1: frames [20480..40960] (first utterance at frame 20480), data subset 0 of 2, with 1 datapasses
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:11: Starting minibatch loop, distributed reading is ENABLED.
MPI Rank 0: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[   1-   3, 3.75%]: CrossEntropyWithSoftmax = 2.21403606 * 508; EvalClassificationError = 0.60039370 * 508; time = 0.3002s; samplesPerSecond = 1692.0
MPI Rank 0: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[   4-   6, 7.50%]: CrossEntropyWithSoftmax = 2.19798626 * 492; EvalClassificationError = 0.58943089 * 492; time = 0.1221s; samplesPerSecond = 4031.0
MPI Rank 0: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[   7-   9, 11.25%]: CrossEntropyWithSoftmax = 2.18333883 * 488; EvalClassificationError = 0.59221311 * 488; time = 0.1245s; samplesPerSecond = 3920.4
MPI Rank 0: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[  10-  12, 15.00%]: CrossEntropyWithSoftmax = 2.23415887 * 527; EvalClassificationError = 0.59203036 * 527; time = 0.1817s; samplesPerSecond = 2901.1
MPI Rank 0: 		(model aggregation stats): 1-th sync point was hit, introducing a 0.03-seconds latency this time; accumulated time on sync point = 0.03 seconds , average latency = 0.03 seconds
MPI Rank 0: 		(model aggregation stats) 1-th sync:     1.35 seconds since last report (0.44 seconds on comm.); 4289 samples processed by 2 workers (2163 by me);
MPI Rank 0: 		(model aggregation stats) 1-th sync: totalThroughput = 3.18k samplesPerSecond , throughputPerWorker = 1.59k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  13-  15, 18.75%]: CrossEntropyWithSoftmax = 2.00557307 * 473; EvalClassificationError = 0.53911205 * 473; time = 0.6993s; samplesPerSecond = 676.4
MPI Rank 0: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  16-  18, 22.50%]: CrossEntropyWithSoftmax = 2.08710611 * 511; EvalClassificationError = 0.54598826 * 511; time = 0.1314s; samplesPerSecond = 3888.9
MPI Rank 0: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  19-  21, 26.25%]: CrossEntropyWithSoftmax = 2.08405445 * 506; EvalClassificationError = 0.54940711 * 506; time = 0.1421s; samplesPerSecond = 3561.7
MPI Rank 0: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  22-  24, 30.00%]: CrossEntropyWithSoftmax = 2.13007119 * 513; EvalClassificationError = 0.54385965 * 513; time = 0.1434s; samplesPerSecond = 3577.1
MPI Rank 0: 		(model aggregation stats): 2-th sync point was hit, introducing a 0.09-seconds latency this time; accumulated time on sync point = 0.12 seconds , average latency = 0.06 seconds
MPI Rank 0: 		(model aggregation stats) 2-th sync:     1.06 seconds since last report (0.15 seconds on comm.); 4253 samples processed by 2 workers (2180 by me);
MPI Rank 0: 		(model aggregation stats) 2-th sync: totalThroughput = 4.03k samplesPerSecond , throughputPerWorker = 2.02k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  25-  27, 33.75%]: CrossEntropyWithSoftmax = 2.07348053 * 489; EvalClassificationError = 0.56032720 * 489; time = 0.5997s; samplesPerSecond = 815.4
MPI Rank 0: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  28-  30, 37.50%]: CrossEntropyWithSoftmax = 2.08147727 * 494; EvalClassificationError = 0.55060729 * 494; time = 0.1242s; samplesPerSecond = 3977.3
MPI Rank 0: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  31-  33, 41.25%]: CrossEntropyWithSoftmax = 2.12020205 * 499; EvalClassificationError = 0.57314629 * 499; time = 0.1263s; samplesPerSecond = 3951.4
MPI Rank 0: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  34-  36, 45.00%]: CrossEntropyWithSoftmax = 2.07338752 * 490; EvalClassificationError = 0.57346939 * 490; time = 0.1239s; samplesPerSecond = 3956.0
MPI Rank 0: 		(model aggregation stats): 3-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.12 seconds , average latency = 0.04 seconds
MPI Rank 0: 		(model aggregation stats) 3-th sync:     1.14 seconds since last report (0.50 seconds on comm.); 4246 samples processed by 2 workers (2144 by me);
MPI Rank 0: 		(model aggregation stats) 3-th sync: totalThroughput = 3.73k samplesPerSecond , throughputPerWorker = 1.86k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  37-  39, 48.75%]: CrossEntropyWithSoftmax = 1.93802618 * 497; EvalClassificationError = 0.51710262 * 497; time = 0.7170s; samplesPerSecond = 693.1
MPI Rank 0: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  40-  42, 52.50%]: CrossEntropyWithSoftmax = 2.14186209 * 492; EvalClassificationError = 0.60162602 * 492; time = 0.1298s; samplesPerSecond = 3789.8
MPI Rank 0: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  43-  45, 56.25%]: CrossEntropyWithSoftmax = 1.91873366 * 508; EvalClassificationError = 0.55118110 * 508; time = 0.1302s; samplesPerSecond = 3902.3
MPI Rank 0: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  46-  48, 60.00%]: CrossEntropyWithSoftmax = 1.98722402 * 503; EvalClassificationError = 0.53081511 * 503; time = 0.1245s; samplesPerSecond = 4041.2
MPI Rank 0: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  49-  51, 63.75%]: CrossEntropyWithSoftmax = 2.09435600 * 470; EvalClassificationError = 0.58723404 * 470; time = 0.1224s; samplesPerSecond = 3838.8
MPI Rank 0: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  52-  54, 67.50%]: CrossEntropyWithSoftmax = 1.92119978 * 494; EvalClassificationError = 0.53643725 * 494; time = 0.1147s; samplesPerSecond = 4306.1
MPI Rank 0: 12/12/2017 15:18:16:  Epoch[ 2 of 5]-Minibatch[  55-  57, 71.25%]: CrossEntropyWithSoftmax = 2.00050314 * 503; EvalClassificationError = 0.53081511 * 503; time = 0.3442s; samplesPerSecond = 1461.5
MPI Rank 0: 12/12/2017 15:18:16:  Epoch[ 2 of 5]-Minibatch[  58-  60, 75.00%]: CrossEntropyWithSoftmax = 1.97136256 * 487; EvalClassificationError = 0.54414784 * 487; time = 0.1105s; samplesPerSecond = 4405.9
MPI Rank 0: 12/12/2017 15:18:16:  Epoch[ 2 of 5]-Minibatch[  61-  63, 78.75%]: CrossEntropyWithSoftmax = 1.97012018 * 516; EvalClassificationError = 0.52713178 * 516; time = 0.1196s; samplesPerSecond = 4314.7
MPI Rank 0: 12/12/2017 15:18:16:  Epoch[ 2 of 5]-Minibatch[  64-  66, 82.50%]: CrossEntropyWithSoftmax = 1.97096226 * 494; EvalClassificationError = 0.54858300 * 494; time = 0.1204s; samplesPerSecond = 4102.0
MPI Rank 0: 12/12/2017 15:18:16:  Epoch[ 2 of 5]-Minibatch[  67-  69, 86.25%]: CrossEntropyWithSoftmax = 2.02297003 * 510; EvalClassificationError = 0.55294118 * 510; time = 0.1172s; samplesPerSecond = 4352.4
MPI Rank 0: 12/12/2017 15:18:16:  Epoch[ 2 of 5]-Minibatch[  70-  72, 90.00%]: CrossEntropyWithSoftmax = 2.03163053 * 497; EvalClassificationError = 0.55935614 * 497; time = 0.1167s; samplesPerSecond = 4260.2
MPI Rank 0: 12/12/2017 15:18:16:  Epoch[ 2 of 5]-Minibatch[  73-  75, 93.75%]: CrossEntropyWithSoftmax = 1.91405684 * 490; EvalClassificationError = 0.55918367 * 490; time = 0.1111s; samplesPerSecond = 4410.8
MPI Rank 0: 12/12/2017 15:18:17:  Epoch[ 2 of 5]-Minibatch[  76-  78, 97.50%]: CrossEntropyWithSoftmax = 1.97528957 * 482; EvalClassificationError = 0.52904564 * 482; time = 0.1222s; samplesPerSecond = 3943.5
MPI Rank 0: 12/12/2017 15:18:17:  Epoch[ 2 of 5]-Minibatch[  79-  81, 101.25%]: CrossEntropyWithSoftmax = 1.99065272 * 342; EvalClassificationError = 0.52339181 * 342; time = 0.0739s; samplesPerSecond = 4626.6
MPI Rank 0: 		(model aggregation stats): 4-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.12 seconds , average latency = 0.03 seconds
MPI Rank 0: 		(model aggregation stats) 4-th sync:     2.07 seconds since last report (0.15 seconds on comm.); 7692 samples processed by 2 workers (6788 by me);
MPI Rank 0: 		(model aggregation stats) 4-th sync: totalThroughput = 3.71k samplesPerSecond , throughputPerWorker = 1.86k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:17: Finished Epoch[ 2 of 5]: [Training] CrossEntropyWithSoftmax = 2.05813627 * 20480; EvalClassificationError = 0.56054688 * 20480; totalSamplesSeen = 40960; learningRatePerSample = 0.001953125; epochTime=5.6295s
MPI Rank 0: 12/12/2017 15:18:17: SGD: Saving checkpoint model '/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/models/cntkSpeech.dnn.2'
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:17: Starting Epoch 3: learning rate per sample = 0.000098  effective momentum = 0.656119  momentum as time constant = 2429.9 samples
MPI Rank 0: Parallel training (2 workers) using BlockMomentumSGD with block momentum = 0.5000, block momentum time constant (per worker) = 2954.6394, block learning rate = 1.0000, block size per worker = 2048 samples, using Nesterov-style block momentum, resetting SGD momentum after sync.
MPI Rank 0: minibatchiterator: epoch 2: frames [40960..61440] (first utterance at frame 40960), data subset 0 of 2, with 1 datapasses
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:17: Starting minibatch loop, distributed reading is ENABLED.
MPI Rank 0: 12/12/2017 15:18:18:  Epoch[ 3 of 5]-Minibatch[   1-   3, 15.00%]: CrossEntropyWithSoftmax = 1.96810913 * 1942; EvalClassificationError = 0.53656025 * 1942; time = 0.5597s; samplesPerSecond = 3469.5
MPI Rank 0: 		(model aggregation stats): 1-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 1-th sync:     1.35 seconds since last report (0.36 seconds on comm.); 4885 samples processed by 2 workers (2592 by me);
MPI Rank 0: 		(model aggregation stats) 1-th sync: totalThroughput = 3.62k samplesPerSecond , throughputPerWorker = 1.81k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:19:  Epoch[ 3 of 5]-Minibatch[   4-   6, 30.00%]: CrossEntropyWithSoftmax = 1.95124060 * 1909; EvalClassificationError = 0.54950236 * 1909; time = 1.0629s; samplesPerSecond = 1796.1
MPI Rank 0: 		(model aggregation stats): 2-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 2-th sync:     1.11 seconds since last report (0.45 seconds on comm.); 4826 samples processed by 2 workers (2577 by me);
MPI Rank 0: 		(model aggregation stats) 2-th sync: totalThroughput = 4.36k samplesPerSecond , throughputPerWorker = 2.18k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:20:  Epoch[ 3 of 5]-Minibatch[   7-   9, 45.00%]: CrossEntropyWithSoftmax = 1.99535428 * 1987; EvalClassificationError = 0.55309512 * 1987; time = 0.9778s; samplesPerSecond = 2032.2
MPI Rank 0: 		(model aggregation stats): 3-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 3-th sync:     0.98 seconds since last report (0.15 seconds on comm.); 4903 samples processed by 2 workers (2577 by me);
MPI Rank 0: 		(model aggregation stats) 3-th sync: totalThroughput = 5.00k samplesPerSecond , throughputPerWorker = 2.50k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:21:  Epoch[ 3 of 5]-Minibatch[  10-  12, 60.00%]: CrossEntropyWithSoftmax = 1.94905959 * 1908; EvalClassificationError = 0.54769392 * 1908; time = 0.8314s; samplesPerSecond = 2295.0
MPI Rank 0: 12/12/2017 15:18:21:  Epoch[ 3 of 5]-Minibatch[  13-  15, 75.00%]: CrossEntropyWithSoftmax = 1.98533312 * 1905; EvalClassificationError = 0.56535433 * 1905; time = 0.4616s; samplesPerSecond = 4127.2
MPI Rank 0: 12/12/2017 15:18:21:  Epoch[ 3 of 5]-Minibatch[  16-  18, 90.00%]: CrossEntropyWithSoftmax = 1.97363277 * 1913; EvalClassificationError = 0.54783063 * 1913; time = 0.4060s; samplesPerSecond = 4711.8
MPI Rank 0: 12/12/2017 15:18:22:  Epoch[ 3 of 5]-Minibatch[  19-  21, 105.00%]: CrossEntropyWithSoftmax = 1.98504692 * 1225; EvalClassificationError = 0.54857143 * 1225; time = 0.2450s; samplesPerSecond = 5000.8
MPI Rank 0: 		(model aggregation stats): 4-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 4-th sync:     1.83 seconds since last report (0.53 seconds on comm.); 5866 samples processed by 2 workers (5043 by me);
MPI Rank 0: 		(model aggregation stats) 4-th sync: totalThroughput = 3.20k samplesPerSecond , throughputPerWorker = 1.60k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:22: Finished Epoch[ 3 of 5]: [Training] CrossEntropyWithSoftmax = 1.96686676 * 20480; EvalClassificationError = 0.54916992 * 20480; totalSamplesSeen = 61440; learningRatePerSample = 9.7656251e-05; epochTime=5.312s
MPI Rank 0: 12/12/2017 15:18:23: SGD: Saving checkpoint model '/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/models/cntkSpeech.dnn.3'
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:23: Starting Epoch 4: learning rate per sample = 0.000098  effective momentum = 0.656119  momentum as time constant = 2429.9 samples
MPI Rank 0: Parallel training (2 workers) using BlockMomentumSGD with block momentum = 0.5000, block momentum time constant (per worker) = 2954.6394, block learning rate = 1.0000, block size per worker = 2048 samples, using Nesterov-style block momentum, resetting SGD momentum after sync.
MPI Rank 0: minibatchiterator: epoch 3: frames [61440..81920] (first utterance at frame 61440), data subset 0 of 2, with 1 datapasses
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:23: Starting minibatch loop, distributed reading is ENABLED.
MPI Rank 0: 12/12/2017 15:18:23:  Epoch[ 4 of 5]-Minibatch[   1-   3, 15.00%]: CrossEntropyWithSoftmax = 1.90230258 * 1923; EvalClassificationError = 0.52418097 * 1923; time = 0.4620s; samplesPerSecond = 4162.6
MPI Rank 0: 		(model aggregation stats): 1-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 1-th sync:     0.92 seconds since last report (0.26 seconds on comm.); 4901 samples processed by 2 workers (2550 by me);
MPI Rank 0: 		(model aggregation stats) 1-th sync: totalThroughput = 5.30k samplesPerSecond , throughputPerWorker = 2.65k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:24:  Epoch[ 4 of 5]-Minibatch[   4-   6, 30.00%]: CrossEntropyWithSoftmax = 1.90656317 * 1870; EvalClassificationError = 0.52566845 * 1870; time = 0.7299s; samplesPerSecond = 2562.1
MPI Rank 0: 		(model aggregation stats): 2-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 2-th sync:     1.02 seconds since last report (0.34 seconds on comm.); 4836 samples processed by 2 workers (2519 by me);
MPI Rank 0: 		(model aggregation stats) 2-th sync: totalThroughput = 4.75k samplesPerSecond , throughputPerWorker = 2.38k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:25:  Epoch[ 4 of 5]-Minibatch[   7-   9, 45.00%]: CrossEntropyWithSoftmax = 1.93014756 * 1942; EvalClassificationError = 0.54325438 * 1942; time = 0.9029s; samplesPerSecond = 2150.9
MPI Rank 0: 		(model aggregation stats): 3-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 3-th sync:     0.84 seconds since last report (0.19 seconds on comm.); 4952 samples processed by 2 workers (2551 by me);
MPI Rank 0: 		(model aggregation stats) 3-th sync: totalThroughput = 5.92k samplesPerSecond , throughputPerWorker = 2.96k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:25:  Epoch[ 4 of 5]-Minibatch[  10-  12, 60.00%]: CrossEntropyWithSoftmax = 1.89401646 * 1885; EvalClassificationError = 0.52625995 * 1885; time = 0.6795s; samplesPerSecond = 2774.1
MPI Rank 0: 12/12/2017 15:18:26:  Epoch[ 4 of 5]-Minibatch[  13-  15, 75.00%]: CrossEntropyWithSoftmax = 1.91437825 * 1870; EvalClassificationError = 0.51978610 * 1870; time = 0.4296s; samplesPerSecond = 4352.9
MPI Rank 0: 12/12/2017 15:18:26:  Epoch[ 4 of 5]-Minibatch[  16-  18, 90.00%]: CrossEntropyWithSoftmax = 1.90290899 * 1873; EvalClassificationError = 0.53230112 * 1873; time = 0.3946s; samplesPerSecond = 4746.6
MPI Rank 0: 12/12/2017 15:18:27:  Epoch[ 4 of 5]-Minibatch[  19-  21, 105.00%]: CrossEntropyWithSoftmax = 1.94263473 * 1231; EvalClassificationError = 0.53533712 * 1231; time = 0.4530s; samplesPerSecond = 2717.7
MPI Rank 0: 		(model aggregation stats): 4-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 4-th sync:     1.53 seconds since last report (0.16 seconds on comm.); 5791 samples processed by 2 workers (4974 by me);
MPI Rank 0: 		(model aggregation stats) 4-th sync: totalThroughput = 3.79k samplesPerSecond , throughputPerWorker = 1.89k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:27: Finished Epoch[ 4 of 5]: [Training] CrossEntropyWithSoftmax = 1.91580675 * 20480; EvalClassificationError = 0.53129883 * 20480; totalSamplesSeen = 81920; learningRatePerSample = 9.7656251e-05; epochTime=4.33567s
MPI Rank 0: 12/12/2017 15:18:27: SGD: Saving checkpoint model '/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/models/cntkSpeech.dnn.4'
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:27: Starting Epoch 5: learning rate per sample = 0.000098  effective momentum = 0.656119  momentum as time constant = 2429.9 samples
MPI Rank 0: Parallel training (2 workers) using BlockMomentumSGD with block momentum = 0.5000, block momentum time constant (per worker) = 2954.6394, block learning rate = 1.0000, block size per worker = 2048 samples, using Nesterov-style block momentum, resetting SGD momentum after sync.
MPI Rank 0: minibatchiterator: epoch 4: frames [81920..102400] (first utterance at frame 81920), data subset 0 of 2, with 1 datapasses
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:27: Starting minibatch loop, distributed reading is ENABLED.
MPI Rank 0: 12/12/2017 15:18:28:  Epoch[ 5 of 5]-Minibatch[   1-   3, 15.00%]: CrossEntropyWithSoftmax = 1.94181066 * 1863; EvalClassificationError = 0.52710682 * 1863; time = 0.4566s; samplesPerSecond = 4080.3
MPI Rank 0: 		(model aggregation stats): 1-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 1-th sync:     1.04 seconds since last report (0.37 seconds on comm.); 4919 samples processed by 2 workers (2493 by me);
MPI Rank 0: 		(model aggregation stats) 1-th sync: totalThroughput = 4.73k samplesPerSecond , throughputPerWorker = 2.36k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:29:  Epoch[ 5 of 5]-Minibatch[   4-   6, 30.00%]: CrossEntropyWithSoftmax = 1.88137918 * 1855; EvalClassificationError = 0.52452830 * 1855; time = 0.8601s; samplesPerSecond = 2156.7
MPI Rank 0: 		(model aggregation stats): 2-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 2-th sync:     1.24 seconds since last report (0.40 seconds on comm.); 4899 samples processed by 2 workers (2480 by me);
MPI Rank 0: 		(model aggregation stats) 2-th sync: totalThroughput = 3.96k samplesPerSecond , throughputPerWorker = 1.98k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:30:  Epoch[ 5 of 5]-Minibatch[   7-   9, 45.00%]: CrossEntropyWithSoftmax = 1.87667090 * 1866; EvalClassificationError = 0.52197213 * 1866; time = 1.1437s; samplesPerSecond = 1631.5
MPI Rank 0: 		(model aggregation stats): 3-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 3-th sync:     0.84 seconds since last report (0.16 seconds on comm.); 4829 samples processed by 2 workers (2470 by me);
MPI Rank 0: 		(model aggregation stats) 3-th sync: totalThroughput = 5.72k samplesPerSecond , throughputPerWorker = 2.86k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:30:  Epoch[ 5 of 5]-Minibatch[  10-  12, 60.00%]: CrossEntropyWithSoftmax = 1.89594264 * 1859; EvalClassificationError = 0.52931684 * 1859; time = 0.6567s; samplesPerSecond = 2830.9
MPI Rank 0: 12/12/2017 15:18:31:  Epoch[ 5 of 5]-Minibatch[  13-  15, 75.00%]: CrossEntropyWithSoftmax = 1.82825460 * 1925; EvalClassificationError = 0.50077922 * 1925; time = 0.7353s; samplesPerSecond = 2618.1
MPI Rank 0: 12/12/2017 15:18:32:  Epoch[ 5 of 5]-Minibatch[  16-  18, 90.00%]: CrossEntropyWithSoftmax = 1.85359611 * 1860; EvalClassificationError = 0.50806452 * 1860; time = 0.6394s; samplesPerSecond = 2908.9
MPI Rank 0: 12/12/2017 15:18:32:  Epoch[ 5 of 5]-Minibatch[  19-  21, 105.00%]: CrossEntropyWithSoftmax = 1.86855308 * 1239; EvalClassificationError = 0.51251009 * 1239; time = 0.2473s; samplesPerSecond = 5009.9
MPI Rank 0: 		(model aggregation stats): 4-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 0: 		(model aggregation stats) 4-th sync:     1.89 seconds since last report (0.21 seconds on comm.); 5833 samples processed by 2 workers (5024 by me);
MPI Rank 0: 		(model aggregation stats) 4-th sync: totalThroughput = 3.08k samplesPerSecond , throughputPerWorker = 1.54k samplesPerSecond
MPI Rank 0: 12/12/2017 15:18:32: Finished Epoch[ 5 of 5]: [Training] CrossEntropyWithSoftmax = 1.89863387 * 20480; EvalClassificationError = 0.52202148 * 20480; totalSamplesSeen = 102400; learningRatePerSample = 9.7656251e-05; epochTime=5.10182s
MPI Rank 0: 12/12/2017 15:18:32: SGD: Saving checkpoint model '/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/models/cntkSpeech.dnn'
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:32: Action "train" complete.
MPI Rank 0: 
MPI Rank 0: 12/12/2017 15:18:32: __COMPLETED__
MPI Rank 1: CNTK 2.3.1+ (HEAD f4f0f8, Dec 11 2017 18:34:12) at 2017/12/12 15:17:58
MPI Rank 1: 
MPI Rank 1: /home/ubuntu/workspace/build/gpu/release/bin/cntk  configFile=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/../ParallelBM/cntk.cntk  currentDirectory=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data  RunDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu  DataDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data  ConfigDir=/home/ubuntu/workspace/Tests/EndToEndTests/Speech/DNN/ParallelBM/..  OutputDir=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu  DeviceId=-1  timestamping=true  numCPUThreads=6  precision=double  speechTrain=[SGD=[ParallelTrain=[parallelizationStartEpoch=2]]]  stderr=/tmp/cntk-test-20171211223423.932710/Speech/DNN_ParallelBM@release_cpu/stderr
MPI Rank 1: 12/12/2017 15:17:58: -------------------------------------------------------------------
MPI Rank 1: 12/12/2017 15:17:58: Build info: 
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:17:58: 		Built time: Dec 11 2017 18:28:39
MPI Rank 1: 12/12/2017 15:17:58: 		Last modified date: Wed Nov 15 09:27:10 2017
MPI Rank 1: 12/12/2017 15:17:58: 		Build type: release
MPI Rank 1: 12/12/2017 15:17:58: 		Build target: GPU
MPI Rank 1: 12/12/2017 15:17:58: 		With ASGD: yes
MPI Rank 1: 12/12/2017 15:17:58: 		Math lib: mkl
MPI Rank 1: 12/12/2017 15:17:58: 		CUDA version: 9.0.0
MPI Rank 1: 12/12/2017 15:17:58: 		CUDNN version: 7.0.4
MPI Rank 1: 12/12/2017 15:17:58: 		Build Branch: HEAD
MPI Rank 1: 12/12/2017 15:17:58: 		Build SHA1: f4f0f82eabcc482dbd03af1f946a44ae2b8b97bf
MPI Rank 1: 12/12/2017 15:17:58: 		MPI distribution: Open MPI
MPI Rank 1: 12/12/2017 15:17:58: 		MPI version: 1.10.7
MPI Rank 1: 12/12/2017 15:17:58: -------------------------------------------------------------------
MPI Rank 1: 12/12/2017 15:17:58: -------------------------------------------------------------------
MPI Rank 1: 12/12/2017 15:17:58: GPU info:
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:17:58: 		Device[0]: cores = 3072; computeCapability = 5.2; type = "Tesla M60"; total memory = 8123 MB; free memory = 8029 MB
MPI Rank 1: 12/12/2017 15:17:58: -------------------------------------------------------------------
MPI Rank 1: 12/12/2017 15:17:58: Using 6 CPU threads.
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:17:58: ##############################################################################
MPI Rank 1: 12/12/2017 15:17:58: #                                                                            #
MPI Rank 1: 12/12/2017 15:17:58: # speechTrain command (train action)                                         #
MPI Rank 1: 12/12/2017 15:17:58: #                                                                            #
MPI Rank 1: 12/12/2017 15:17:58: ##############################################################################
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:17:58: 
MPI Rank 1: Creating virgin network.
MPI Rank 1: SimpleNetworkBuilder Using CPU
MPI Rank 1: reading script file glob_0000.scp ... 948 entries
MPI Rank 1: total 132 state names in state list /home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data/state.list
MPI Rank 1: htkmlfreader: reading MLF file /home/ubuntu/workspace/Tests/EndToEndTests/Speech/Data/glob_0000.mlf ... total 948 entries
MPI Rank 1: ...............................................................................................feature set 0: 252734 frames in 948 out of 948 utterances
MPI Rank 1: label set 0: 129 classes
MPI Rank 1: minibatchutterancesource: 948 utterances grouped into 3 chunks, av. chunk size: 316.0 utterances, 84244.7 frames
MPI Rank 1: 12/12/2017 15:17:58: 
MPI Rank 1: Model has 25 nodes. Using CPU.
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:17:58: Training criterion:   CrossEntropyWithSoftmax = CrossEntropyWithSoftmax
MPI Rank 1: 12/12/2017 15:17:58: Evaluation criterion: EvalClassificationError = ClassificationError
MPI Rank 1: 
MPI Rank 1: 
MPI Rank 1: Allocating matrices for forward and/or backward propagation.
MPI Rank 1: 
MPI Rank 1: Gradient Memory Aliasing: 4 are aliased.
MPI Rank 1: 	W2*H1 (gradient) reuses HLast (gradient)
MPI Rank 1: 	W1*H1 (gradient) reuses W1*H1+B1 (gradient)
MPI Rank 1: 
MPI Rank 1: Memory Sharing: Out of 40 matrices, 21 are shared as 5, and 19 are not shared.
MPI Rank 1: 
MPI Rank 1: Here are the ones that share memory:
MPI Rank 1: 	{ PosteriorProb : [132 x 1 x *]
MPI Rank 1: 	  ScaledLogLikelihood : [132 x 1 x *] }
MPI Rank 1: 	{ H2 : [512 x 1 x *]
MPI Rank 1: 	  W0*features+B0 : [512 x 1 x *]
MPI Rank 1: 	  W1 : [512 x 512] (gradient)
MPI Rank 1: 	  W1*H1 : [512 x 1 x *] }
MPI Rank 1: 	{ H1 : [512 x 1 x *] (gradient)
MPI Rank 1: 	  H2 : [512 x 1 x *] (gradient)
MPI Rank 1: 	  HLast : [132 x 1 x *]
MPI Rank 1: 	  W0*features : [512 x *]
MPI Rank 1: 	  W0*features : [512 x *] (gradient) }
MPI Rank 1: 	{ HLast : [132 x 1 x *] (gradient)
MPI Rank 1: 	  W0 : [512 x 363] (gradient)
MPI Rank 1: 	  W0*features+B0 : [512 x 1 x *] (gradient)
MPI Rank 1: 	  W1*H1 : [512 x 1 x *] (gradient)
MPI Rank 1: 	  W1*H1+B1 : [512 x 1 x *]
MPI Rank 1: 	  W1*H1+B1 : [512 x 1 x *] (gradient)
MPI Rank 1: 	  W2*H1 : [132 x 1 x *]
MPI Rank 1: 	  W2*H1 : [132 x 1 x *] (gradient) }
MPI Rank 1: 	{ B0 : [512 x 1] (gradient)
MPI Rank 1: 	  H1 : [512 x 1 x *] }
MPI Rank 1: 
MPI Rank 1: Here are the ones that don't share memory:
MPI Rank 1: 	{features : [363 x *]}
MPI Rank 1: 	{InvStdOfFeatures : [363]}
MPI Rank 1: 	{MeanOfFeatures : [363]}
MPI Rank 1: 	{W0 : [512 x 363]}
MPI Rank 1: 	{B0 : [512 x 1]}
MPI Rank 1: 	{W1 : [512 x 512]}
MPI Rank 1: 	{B1 : [512 x 1]}
MPI Rank 1: 	{W2 : [132 x 512]}
MPI Rank 1: 	{B2 : [132 x 1]}
MPI Rank 1: 	{labels : [132 x *]}
MPI Rank 1: 	{Prior : [132]}
MPI Rank 1: 	{CrossEntropyWithSoftmax : [1]}
MPI Rank 1: 	{EvalClassificationError : [1]}
MPI Rank 1: 	{B1 : [512 x 1] (gradient)}
MPI Rank 1: 	{MVNormalizedFeatures : [363 x *]}
MPI Rank 1: 	{W2 : [132 x 512] (gradient)}
MPI Rank 1: 	{B2 : [132 x 1] (gradient)}
MPI Rank 1: 	{LogOfPrior : [132]}
MPI Rank 1: 	{CrossEntropyWithSoftmax : [1] (gradient)}
MPI Rank 1: 
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:17:58: Training 516740 parameters in 6 out of 6 parameter tensors and 15 nodes with gradient:
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:17:58: 	Node 'B0' (LearnableParameter operation) : [512 x 1]
MPI Rank 1: 12/12/2017 15:17:58: 	Node 'B1' (LearnableParameter operation) : [512 x 1]
MPI Rank 1: 12/12/2017 15:17:58: 	Node 'B2' (LearnableParameter operation) : [132 x 1]
MPI Rank 1: 12/12/2017 15:17:58: 	Node 'W0' (LearnableParameter operation) : [512 x 363]
MPI Rank 1: 12/12/2017 15:17:58: 	Node 'W1' (LearnableParameter operation) : [512 x 512]
MPI Rank 1: 12/12/2017 15:17:58: 	Node 'W2' (LearnableParameter operation) : [132 x 512]
MPI Rank 1: 
MPI Rank 1: NcclComm: disabled, at least one rank using CPU device
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:17:59: Precomputing --> 3 PreCompute nodes found.
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:17:59: 	MeanOfFeatures = Mean()
MPI Rank 1: 12/12/2017 15:17:59: 	InvStdOfFeatures = InvStdDev()
MPI Rank 1: 12/12/2017 15:17:59: 	Prior = Mean()
MPI Rank 1: minibatchiterator: epoch 0: frames [0..252734] (first utterance at frame 0), data subset 0 of 1, with 1 datapasses
MPI Rank 1: requiredata: determined feature kind as 33-dimensional 'USER' with frame shift 10.0 ms
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:04: Precomputing --> Completed.
MPI Rank 1: 
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:04: Starting Epoch 1: learning rate per sample = 0.015625  effective momentum = 0.900000  momentum as time constant = 607.4 samples
MPI Rank 1: minibatchiterator: epoch 0: frames [0..20480] (first utterance at frame 0), data subset 0 of 1, with 1 datapasses
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:04: Starting minibatch loop.
MPI Rank 1: 12/12/2017 15:18:04:  Epoch[ 1 of 5]-Minibatch[   1-   3, 0.94%]: CrossEntropyWithSoftmax = 4.65054456 * 192; EvalClassificationError = 0.96875000 * 192; time = 0.0915s; samplesPerSecond = 2097.6
MPI Rank 1: 12/12/2017 15:18:04:  Epoch[ 1 of 5]-Minibatch[   4-   6, 1.88%]: CrossEntropyWithSoftmax = 4.41918514 * 192; EvalClassificationError = 0.88020833 * 192; time = 0.0582s; samplesPerSecond = 3299.7
MPI Rank 1: 12/12/2017 15:18:04:  Epoch[ 1 of 5]-Minibatch[   7-   9, 2.81%]: CrossEntropyWithSoftmax = 4.76293439 * 192; EvalClassificationError = 0.93229167 * 192; time = 0.0573s; samplesPerSecond = 3351.1
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  10-  12, 3.75%]: CrossEntropyWithSoftmax = 4.23168138 * 192; EvalClassificationError = 0.90104167 * 192; time = 0.0593s; samplesPerSecond = 3240.1
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  13-  15, 4.69%]: CrossEntropyWithSoftmax = 4.47386985 * 192; EvalClassificationError = 0.89583333 * 192; time = 0.0580s; samplesPerSecond = 3312.4
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  16-  18, 5.62%]: CrossEntropyWithSoftmax = 4.39795080 * 192; EvalClassificationError = 0.94270833 * 192; time = 0.0567s; samplesPerSecond = 3387.2
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  19-  21, 6.56%]: CrossEntropyWithSoftmax = 4.25148851 * 192; EvalClassificationError = 0.97395833 * 192; time = 0.0561s; samplesPerSecond = 3424.8
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  22-  24, 7.50%]: CrossEntropyWithSoftmax = 4.06906673 * 192; EvalClassificationError = 0.89062500 * 192; time = 0.2594s; samplesPerSecond = 740.1
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  25-  27, 8.44%]: CrossEntropyWithSoftmax = 3.98165780 * 192; EvalClassificationError = 0.90104167 * 192; time = 0.0556s; samplesPerSecond = 3450.8
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  28-  30, 9.38%]: CrossEntropyWithSoftmax = 3.84788960 * 192; EvalClassificationError = 0.85416667 * 192; time = 0.0560s; samplesPerSecond = 3428.3
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  31-  33, 10.31%]: CrossEntropyWithSoftmax = 3.80540005 * 192; EvalClassificationError = 0.84895833 * 192; time = 0.0564s; samplesPerSecond = 3405.7
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  34-  36, 11.25%]: CrossEntropyWithSoftmax = 3.76164351 * 192; EvalClassificationError = 0.86979167 * 192; time = 0.0555s; samplesPerSecond = 3462.1
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  37-  39, 12.19%]: CrossEntropyWithSoftmax = 3.74042928 * 192; EvalClassificationError = 0.82812500 * 192; time = 0.0571s; samplesPerSecond = 3360.5
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  40-  42, 13.12%]: CrossEntropyWithSoftmax = 3.69572235 * 192; EvalClassificationError = 0.83333333 * 192; time = 0.0563s; samplesPerSecond = 3410.3
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  43-  45, 14.06%]: CrossEntropyWithSoftmax = 3.89722237 * 192; EvalClassificationError = 0.90104167 * 192; time = 0.0563s; samplesPerSecond = 3413.1
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  46-  48, 15.00%]: CrossEntropyWithSoftmax = 3.77423768 * 192; EvalClassificationError = 0.84375000 * 192; time = 0.0564s; samplesPerSecond = 3405.3
MPI Rank 1: 12/12/2017 15:18:05:  Epoch[ 1 of 5]-Minibatch[  49-  51, 15.94%]: CrossEntropyWithSoftmax = 3.81262710 * 192; EvalClassificationError = 0.88541667 * 192; time = 0.0570s; samplesPerSecond = 3370.3
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  52-  54, 16.88%]: CrossEntropyWithSoftmax = 3.92558806 * 192; EvalClassificationError = 0.88020833 * 192; time = 0.0586s; samplesPerSecond = 3278.0
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  55-  57, 17.81%]: CrossEntropyWithSoftmax = 3.56995707 * 192; EvalClassificationError = 0.84895833 * 192; time = 0.0556s; samplesPerSecond = 3452.0
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  58-  60, 18.75%]: CrossEntropyWithSoftmax = 3.67092670 * 192; EvalClassificationError = 0.88541667 * 192; time = 0.0553s; samplesPerSecond = 3469.4
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  61-  63, 19.69%]: CrossEntropyWithSoftmax = 3.36174584 * 192; EvalClassificationError = 0.77083333 * 192; time = 0.0590s; samplesPerSecond = 3252.6
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  64-  66, 20.62%]: CrossEntropyWithSoftmax = 3.42123462 * 192; EvalClassificationError = 0.81770833 * 192; time = 0.0542s; samplesPerSecond = 3541.5
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  67-  69, 21.56%]: CrossEntropyWithSoftmax = 3.36684019 * 192; EvalClassificationError = 0.79166667 * 192; time = 0.0556s; samplesPerSecond = 3451.7
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  70-  72, 22.50%]: CrossEntropyWithSoftmax = 3.68558416 * 192; EvalClassificationError = 0.84375000 * 192; time = 0.0566s; samplesPerSecond = 3392.7
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  73-  75, 23.44%]: CrossEntropyWithSoftmax = 3.64969026 * 192; EvalClassificationError = 0.82812500 * 192; time = 0.0569s; samplesPerSecond = 3371.7
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  76-  78, 24.38%]: CrossEntropyWithSoftmax = 3.37812633 * 192; EvalClassificationError = 0.81770833 * 192; time = 0.0549s; samplesPerSecond = 3497.8
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  79-  81, 25.31%]: CrossEntropyWithSoftmax = 3.52781334 * 192; EvalClassificationError = 0.82291667 * 192; time = 0.0554s; samplesPerSecond = 3465.6
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  82-  84, 26.25%]: CrossEntropyWithSoftmax = 3.47704383 * 192; EvalClassificationError = 0.81250000 * 192; time = 0.0606s; samplesPerSecond = 3169.1
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  85-  87, 27.19%]: CrossEntropyWithSoftmax = 3.56621904 * 192; EvalClassificationError = 0.82812500 * 192; time = 0.0563s; samplesPerSecond = 3409.3
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  88-  90, 28.12%]: CrossEntropyWithSoftmax = 3.44116285 * 192; EvalClassificationError = 0.80729167 * 192; time = 0.0559s; samplesPerSecond = 3432.9
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  91-  93, 29.06%]: CrossEntropyWithSoftmax = 3.37856471 * 192; EvalClassificationError = 0.80729167 * 192; time = 0.0590s; samplesPerSecond = 3255.8
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  94-  96, 30.00%]: CrossEntropyWithSoftmax = 3.56985531 * 192; EvalClassificationError = 0.82291667 * 192; time = 0.0588s; samplesPerSecond = 3265.7
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[  97-  99, 30.94%]: CrossEntropyWithSoftmax = 3.36523472 * 192; EvalClassificationError = 0.81770833 * 192; time = 0.0593s; samplesPerSecond = 3236.2
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[ 100- 102, 31.87%]: CrossEntropyWithSoftmax = 3.51274161 * 192; EvalClassificationError = 0.82291667 * 192; time = 0.0554s; samplesPerSecond = 3467.9
MPI Rank 1: 12/12/2017 15:18:06:  Epoch[ 1 of 5]-Minibatch[ 103- 105, 32.81%]: CrossEntropyWithSoftmax = 3.57161359 * 192; EvalClassificationError = 0.83333333 * 192; time = 0.0572s; samplesPerSecond = 3356.2
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 106- 108, 33.75%]: CrossEntropyWithSoftmax = 3.50343297 * 192; EvalClassificationError = 0.82812500 * 192; time = 0.0623s; samplesPerSecond = 3083.6
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 109- 111, 34.69%]: CrossEntropyWithSoftmax = 3.34024400 * 192; EvalClassificationError = 0.79687500 * 192; time = 0.0580s; samplesPerSecond = 3311.5
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 112- 114, 35.62%]: CrossEntropyWithSoftmax = 3.30238305 * 192; EvalClassificationError = 0.80729167 * 192; time = 0.0591s; samplesPerSecond = 3251.4
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 115- 117, 36.56%]: CrossEntropyWithSoftmax = 3.21012106 * 192; EvalClassificationError = 0.78645833 * 192; time = 0.0573s; samplesPerSecond = 3352.3
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 118- 120, 37.50%]: CrossEntropyWithSoftmax = 3.27497388 * 192; EvalClassificationError = 0.77604167 * 192; time = 0.0574s; samplesPerSecond = 3344.8
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 121- 123, 38.44%]: CrossEntropyWithSoftmax = 3.55026166 * 192; EvalClassificationError = 0.83854167 * 192; time = 0.0603s; samplesPerSecond = 3181.6
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 124- 126, 39.38%]: CrossEntropyWithSoftmax = 3.01866034 * 192; EvalClassificationError = 0.73958333 * 192; time = 0.2610s; samplesPerSecond = 735.6
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 127- 129, 40.31%]: CrossEntropyWithSoftmax = 3.13818688 * 192; EvalClassificationError = 0.80208333 * 192; time = 0.0552s; samplesPerSecond = 3478.1
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 130- 132, 41.25%]: CrossEntropyWithSoftmax = 3.07582410 * 192; EvalClassificationError = 0.76041667 * 192; time = 0.0599s; samplesPerSecond = 3206.1
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 133- 135, 42.19%]: CrossEntropyWithSoftmax = 3.19786112 * 192; EvalClassificationError = 0.75520833 * 192; time = 0.0592s; samplesPerSecond = 3242.7
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 136- 138, 43.12%]: CrossEntropyWithSoftmax = 2.79776977 * 192; EvalClassificationError = 0.71875000 * 192; time = 0.0585s; samplesPerSecond = 3284.1
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 139- 141, 44.06%]: CrossEntropyWithSoftmax = 3.19303582 * 192; EvalClassificationError = 0.82291667 * 192; time = 0.0562s; samplesPerSecond = 3413.8
MPI Rank 1: 12/12/2017 15:18:07:  Epoch[ 1 of 5]-Minibatch[ 142- 144, 45.00%]: CrossEntropyWithSoftmax = 3.04555903 * 192; EvalClassificationError = 0.75520833 * 192; time = 0.0568s; samplesPerSecond = 3378.9
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 145- 147, 45.94%]: CrossEntropyWithSoftmax = 3.03596679 * 192; EvalClassificationError = 0.72916667 * 192; time = 0.0623s; samplesPerSecond = 3083.0
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 148- 150, 46.88%]: CrossEntropyWithSoftmax = 2.81093303 * 192; EvalClassificationError = 0.64583333 * 192; time = 0.0569s; samplesPerSecond = 3376.4
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 151- 153, 47.81%]: CrossEntropyWithSoftmax = 3.05364288 * 192; EvalClassificationError = 0.69791667 * 192; time = 0.0613s; samplesPerSecond = 3130.2
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 154- 156, 48.75%]: CrossEntropyWithSoftmax = 3.18200713 * 192; EvalClassificationError = 0.78645833 * 192; time = 0.0576s; samplesPerSecond = 3333.0
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 157- 159, 49.69%]: CrossEntropyWithSoftmax = 3.13461023 * 192; EvalClassificationError = 0.75000000 * 192; time = 0.0577s; samplesPerSecond = 3329.9
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 160- 162, 50.62%]: CrossEntropyWithSoftmax = 2.88863478 * 192; EvalClassificationError = 0.72395833 * 192; time = 0.0541s; samplesPerSecond = 3549.3
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 163- 165, 51.56%]: CrossEntropyWithSoftmax = 2.81095072 * 192; EvalClassificationError = 0.70833333 * 192; time = 0.0552s; samplesPerSecond = 3476.1
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 166- 168, 52.50%]: CrossEntropyWithSoftmax = 3.02416042 * 192; EvalClassificationError = 0.73958333 * 192; time = 0.0562s; samplesPerSecond = 3418.2
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 169- 171, 53.44%]: CrossEntropyWithSoftmax = 2.75149996 * 192; EvalClassificationError = 0.69791667 * 192; time = 0.0557s; samplesPerSecond = 3444.2
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 172- 174, 54.37%]: CrossEntropyWithSoftmax = 2.66519087 * 192; EvalClassificationError = 0.65625000 * 192; time = 0.0548s; samplesPerSecond = 3505.9
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 175- 177, 55.31%]: CrossEntropyWithSoftmax = 2.73446036 * 192; EvalClassificationError = 0.65625000 * 192; time = 0.0590s; samplesPerSecond = 3253.7
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 178- 180, 56.25%]: CrossEntropyWithSoftmax = 2.76060156 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0550s; samplesPerSecond = 3491.1
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 181- 183, 57.19%]: CrossEntropyWithSoftmax = 2.84893104 * 192; EvalClassificationError = 0.73958333 * 192; time = 0.0562s; samplesPerSecond = 3419.3
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 184- 186, 58.13%]: CrossEntropyWithSoftmax = 2.82041431 * 192; EvalClassificationError = 0.72916667 * 192; time = 0.0592s; samplesPerSecond = 3241.1
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 187- 189, 59.06%]: CrossEntropyWithSoftmax = 2.81359166 * 192; EvalClassificationError = 0.72916667 * 192; time = 0.0587s; samplesPerSecond = 3272.1
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 190- 192, 60.00%]: CrossEntropyWithSoftmax = 2.80548960 * 192; EvalClassificationError = 0.71354167 * 192; time = 0.0551s; samplesPerSecond = 3486.5
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 193- 195, 60.94%]: CrossEntropyWithSoftmax = 2.57693072 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0553s; samplesPerSecond = 3472.6
MPI Rank 1: 12/12/2017 15:18:08:  Epoch[ 1 of 5]-Minibatch[ 196- 198, 61.88%]: CrossEntropyWithSoftmax = 3.00545481 * 192; EvalClassificationError = 0.74479167 * 192; time = 0.0597s; samplesPerSecond = 3218.7
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 199- 201, 62.81%]: CrossEntropyWithSoftmax = 2.66074209 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0559s; samplesPerSecond = 3432.9
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 202- 204, 63.75%]: CrossEntropyWithSoftmax = 2.55576220 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0580s; samplesPerSecond = 3309.8
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 205- 207, 64.69%]: CrossEntropyWithSoftmax = 2.63325580 * 192; EvalClassificationError = 0.66666667 * 192; time = 0.0539s; samplesPerSecond = 3564.0
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 208- 210, 65.62%]: CrossEntropyWithSoftmax = 2.59688083 * 192; EvalClassificationError = 0.67187500 * 192; time = 0.0587s; samplesPerSecond = 3269.6
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 211- 213, 66.56%]: CrossEntropyWithSoftmax = 2.59744697 * 192; EvalClassificationError = 0.61979167 * 192; time = 0.0564s; samplesPerSecond = 3401.9
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 214- 216, 67.50%]: CrossEntropyWithSoftmax = 2.49590166 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0552s; samplesPerSecond = 3475.9
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 217- 219, 68.44%]: CrossEntropyWithSoftmax = 2.75606293 * 192; EvalClassificationError = 0.67708333 * 192; time = 0.0578s; samplesPerSecond = 3322.9
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 220- 222, 69.38%]: CrossEntropyWithSoftmax = 2.51043915 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0575s; samplesPerSecond = 3338.1
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 223- 225, 70.31%]: CrossEntropyWithSoftmax = 2.46191162 * 192; EvalClassificationError = 0.66145833 * 192; time = 0.0563s; samplesPerSecond = 3410.2
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 226- 228, 71.25%]: CrossEntropyWithSoftmax = 2.74930663 * 192; EvalClassificationError = 0.70312500 * 192; time = 0.3092s; samplesPerSecond = 620.9
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 229- 231, 72.19%]: CrossEntropyWithSoftmax = 2.56948343 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0569s; samplesPerSecond = 3372.8
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 232- 234, 73.12%]: CrossEntropyWithSoftmax = 2.99801669 * 192; EvalClassificationError = 0.77083333 * 192; time = 0.0556s; samplesPerSecond = 3456.1
MPI Rank 1: 12/12/2017 15:18:09:  Epoch[ 1 of 5]-Minibatch[ 235- 237, 74.06%]: CrossEntropyWithSoftmax = 2.37847319 * 192; EvalClassificationError = 0.59895833 * 192; time = 0.0626s; samplesPerSecond = 3064.8
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 238- 240, 75.00%]: CrossEntropyWithSoftmax = 2.26592694 * 192; EvalClassificationError = 0.61979167 * 192; time = 0.0562s; samplesPerSecond = 3416.7
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 241- 243, 75.94%]: CrossEntropyWithSoftmax = 2.30186997 * 192; EvalClassificationError = 0.57291667 * 192; time = 0.0563s; samplesPerSecond = 3412.0
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 244- 246, 76.88%]: CrossEntropyWithSoftmax = 2.70240793 * 192; EvalClassificationError = 0.70312500 * 192; time = 0.0604s; samplesPerSecond = 3178.5
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 247- 249, 77.81%]: CrossEntropyWithSoftmax = 2.43935470 * 192; EvalClassificationError = 0.60937500 * 192; time = 0.0590s; samplesPerSecond = 3252.6
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 250- 252, 78.75%]: CrossEntropyWithSoftmax = 2.52037152 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0543s; samplesPerSecond = 3534.2
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 253- 255, 79.69%]: CrossEntropyWithSoftmax = 2.38274509 * 192; EvalClassificationError = 0.63020833 * 192; time = 0.0541s; samplesPerSecond = 3552.0
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 256- 258, 80.62%]: CrossEntropyWithSoftmax = 2.36861217 * 192; EvalClassificationError = 0.59375000 * 192; time = 0.0561s; samplesPerSecond = 3424.1
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 259- 261, 81.56%]: CrossEntropyWithSoftmax = 2.34453624 * 192; EvalClassificationError = 0.63541667 * 192; time = 0.0567s; samplesPerSecond = 3385.1
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 262- 264, 82.50%]: CrossEntropyWithSoftmax = 2.29446007 * 192; EvalClassificationError = 0.59895833 * 192; time = 0.0534s; samplesPerSecond = 3593.6
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 265- 267, 83.44%]: CrossEntropyWithSoftmax = 2.09108193 * 192; EvalClassificationError = 0.52604167 * 192; time = 0.0942s; samplesPerSecond = 2037.6
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 268- 270, 84.38%]: CrossEntropyWithSoftmax = 2.42635931 * 192; EvalClassificationError = 0.66145833 * 192; time = 0.0660s; samplesPerSecond = 2907.4
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 271- 273, 85.31%]: CrossEntropyWithSoftmax = 2.53475002 * 192; EvalClassificationError = 0.66145833 * 192; time = 0.0580s; samplesPerSecond = 3312.2
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 274- 276, 86.25%]: CrossEntropyWithSoftmax = 2.35308728 * 192; EvalClassificationError = 0.61458333 * 192; time = 0.0583s; samplesPerSecond = 3295.2
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 277- 279, 87.19%]: CrossEntropyWithSoftmax = 2.62700347 * 192; EvalClassificationError = 0.69270833 * 192; time = 0.0568s; samplesPerSecond = 3379.6
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 280- 282, 88.12%]: CrossEntropyWithSoftmax = 2.48479326 * 192; EvalClassificationError = 0.61458333 * 192; time = 0.0538s; samplesPerSecond = 3566.5
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 283- 285, 89.06%]: CrossEntropyWithSoftmax = 2.31729432 * 192; EvalClassificationError = 0.62500000 * 192; time = 0.0543s; samplesPerSecond = 3537.8
MPI Rank 1: 12/12/2017 15:18:10:  Epoch[ 1 of 5]-Minibatch[ 286- 288, 90.00%]: CrossEntropyWithSoftmax = 2.21677298 * 192; EvalClassificationError = 0.58854167 * 192; time = 0.0584s; samplesPerSecond = 3288.2
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 289- 291, 90.94%]: CrossEntropyWithSoftmax = 2.28114207 * 192; EvalClassificationError = 0.58333333 * 192; time = 0.0547s; samplesPerSecond = 3507.3
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 292- 294, 91.88%]: CrossEntropyWithSoftmax = 2.12055459 * 192; EvalClassificationError = 0.57812500 * 192; time = 0.0839s; samplesPerSecond = 2288.9
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 295- 297, 92.81%]: CrossEntropyWithSoftmax = 2.31557649 * 192; EvalClassificationError = 0.64062500 * 192; time = 0.0649s; samplesPerSecond = 2958.4
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 298- 300, 93.75%]: CrossEntropyWithSoftmax = 2.30559259 * 192; EvalClassificationError = 0.60937500 * 192; time = 0.0532s; samplesPerSecond = 3610.1
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 301- 303, 94.69%]: CrossEntropyWithSoftmax = 2.15644939 * 192; EvalClassificationError = 0.58854167 * 192; time = 0.0534s; samplesPerSecond = 3593.0
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 304- 306, 95.62%]: CrossEntropyWithSoftmax = 2.26862225 * 192; EvalClassificationError = 0.58333333 * 192; time = 0.0547s; samplesPerSecond = 3509.6
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 307- 309, 96.56%]: CrossEntropyWithSoftmax = 2.15494679 * 192; EvalClassificationError = 0.54166667 * 192; time = 0.0580s; samplesPerSecond = 3312.1
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 310- 312, 97.50%]: CrossEntropyWithSoftmax = 2.44588898 * 192; EvalClassificationError = 0.65104167 * 192; time = 0.0552s; samplesPerSecond = 3478.1
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 313- 315, 98.44%]: CrossEntropyWithSoftmax = 2.17852500 * 192; EvalClassificationError = 0.58854167 * 192; time = 0.0543s; samplesPerSecond = 3533.5
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 1 of 5]-Minibatch[ 316- 318, 99.38%]: CrossEntropyWithSoftmax = 2.23007043 * 192; EvalClassificationError = 0.56770833 * 192; time = 0.0568s; samplesPerSecond = 3381.3
MPI Rank 1: 12/12/2017 15:18:11: Finished Epoch[ 1 of 5]: [Training] CrossEntropyWithSoftmax = 3.04696987 * 20480; EvalClassificationError = 0.73583984 * 20480; totalSamplesSeen = 20480; learningRatePerSample = 0.015625; epochTime=6.85407s
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:11: Starting Epoch 2: learning rate per sample = 0.001953  effective momentum = 0.656119  momentum as time constant = 607.5 samples
MPI Rank 1: Parallel training (2 workers) using BlockMomentumSGD with block momentum = 0.5000, block momentum time constant (per worker) = 2954.6394, block learning rate = 1.0000, block size per worker = 2048 samples, using Nesterov-style block momentum, resetting SGD momentum after sync.
MPI Rank 1: minibatchiterator: epoch 1: frames [20480..40960] (first utterance at frame 20480), data subset 1 of 2, with 1 datapasses
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:11: Starting minibatch loop, distributed reading is ENABLED.
MPI Rank 1: 12/12/2017 15:18:11:  Epoch[ 2 of 5]-Minibatch[   1-   3, 3.75%]: CrossEntropyWithSoftmax = 2.20302741 * 260; EvalClassificationError = 0.60384615 * 260; time = 0.2171s; samplesPerSecond = 1197.8
MPI Rank 1: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[   4-   6, 7.50%]: CrossEntropyWithSoftmax = 2.36629206 * 276; EvalClassificationError = 0.67753623 * 276; time = 0.0842s; samplesPerSecond = 3277.9
MPI Rank 1: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[   7-   9, 11.25%]: CrossEntropyWithSoftmax = 2.25051362 * 280; EvalClassificationError = 0.64642857 * 280; time = 0.0846s; samplesPerSecond = 3310.6
MPI Rank 1: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[  10-  12, 15.00%]: CrossEntropyWithSoftmax = 2.21098962 * 241; EvalClassificationError = 0.64730290 * 241; time = 0.0743s; samplesPerSecond = 3243.1
MPI Rank 1: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[  13-  15, 18.75%]: CrossEntropyWithSoftmax = 2.05433280 * 295; EvalClassificationError = 0.53898305 * 295; time = 0.0848s; samplesPerSecond = 3480.7
MPI Rank 1: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[  16-  18, 22.50%]: CrossEntropyWithSoftmax = 2.37873293 * 257; EvalClassificationError = 0.62256809 * 257; time = 0.0732s; samplesPerSecond = 3509.5
MPI Rank 1: 12/12/2017 15:18:12:  Epoch[ 2 of 5]-Minibatch[  19-  21, 26.25%]: CrossEntropyWithSoftmax = 2.04679577 * 262; EvalClassificationError = 0.56488550 * 262; time = 0.0727s; samplesPerSecond = 3605.9
MPI Rank 1: 		(model aggregation stats): 1-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 1: 		(model aggregation stats) 1-th sync:     1.36 seconds since last report (0.51 seconds on comm.); 4289 samples processed by 2 workers (2126 by me);
MPI Rank 1: 		(model aggregation stats) 1-th sync: totalThroughput = 3.14k samplesPerSecond , throughputPerWorker = 1.57k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  22-  24, 30.00%]: CrossEntropyWithSoftmax = 2.04123164 * 255; EvalClassificationError = 0.52941176 * 255; time = 0.6438s; samplesPerSecond = 396.1
MPI Rank 1: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  25-  27, 33.75%]: CrossEntropyWithSoftmax = 2.08778269 * 279; EvalClassificationError = 0.60215054 * 279; time = 0.0863s; samplesPerSecond = 3234.1
MPI Rank 1: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  28-  30, 37.50%]: CrossEntropyWithSoftmax = 2.05594555 * 274; EvalClassificationError = 0.53649635 * 274; time = 0.0728s; samplesPerSecond = 3762.8
MPI Rank 1: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  31-  33, 41.25%]: CrossEntropyWithSoftmax = 1.95868599 * 269; EvalClassificationError = 0.55018587 * 269; time = 0.0757s; samplesPerSecond = 3554.4
MPI Rank 1: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  34-  36, 45.00%]: CrossEntropyWithSoftmax = 2.04919096 * 278; EvalClassificationError = 0.60791367 * 278; time = 0.1274s; samplesPerSecond = 2181.8
MPI Rank 1: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  37-  39, 48.75%]: CrossEntropyWithSoftmax = 2.13610173 * 271; EvalClassificationError = 0.53505535 * 271; time = 0.0806s; samplesPerSecond = 3361.4
MPI Rank 1: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  40-  42, 52.50%]: CrossEntropyWithSoftmax = 2.21631200 * 276; EvalClassificationError = 0.59057971 * 276; time = 0.0772s; samplesPerSecond = 3576.3
MPI Rank 1: 12/12/2017 15:18:13:  Epoch[ 2 of 5]-Minibatch[  43-  45, 56.25%]: CrossEntropyWithSoftmax = 2.16223397 * 260; EvalClassificationError = 0.57307692 * 260; time = 0.0798s; samplesPerSecond = 3259.9
MPI Rank 1: 		(model aggregation stats): 2-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 1: 		(model aggregation stats) 2-th sync:     1.04 seconds since last report (0.14 seconds on comm.); 4253 samples processed by 2 workers (2073 by me);
MPI Rank 1: 		(model aggregation stats) 2-th sync: totalThroughput = 4.08k samplesPerSecond , throughputPerWorker = 2.04k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  46-  48, 60.00%]: CrossEntropyWithSoftmax = 1.98815673 * 265; EvalClassificationError = 0.56226415 * 265; time = 0.4735s; samplesPerSecond = 559.6
MPI Rank 1: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  49-  51, 63.75%]: CrossEntropyWithSoftmax = 2.13009666 * 298; EvalClassificationError = 0.56711409 * 298; time = 0.0824s; samplesPerSecond = 3618.5
MPI Rank 1: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  52-  54, 67.50%]: CrossEntropyWithSoftmax = 1.97493179 * 274; EvalClassificationError = 0.51459854 * 274; time = 0.0751s; samplesPerSecond = 3646.9
MPI Rank 1: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  55-  57, 71.25%]: CrossEntropyWithSoftmax = 1.98585123 * 265; EvalClassificationError = 0.55094340 * 265; time = 0.0782s; samplesPerSecond = 3388.5
MPI Rank 1: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  58-  60, 75.00%]: CrossEntropyWithSoftmax = 1.85715774 * 281; EvalClassificationError = 0.56227758 * 281; time = 0.0796s; samplesPerSecond = 3529.1
MPI Rank 1: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  61-  63, 78.75%]: CrossEntropyWithSoftmax = 1.95362121 * 252; EvalClassificationError = 0.52777778 * 252; time = 0.0698s; samplesPerSecond = 3609.7
MPI Rank 1: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  64-  66, 82.50%]: CrossEntropyWithSoftmax = 2.04753556 * 274; EvalClassificationError = 0.53649635 * 274; time = 0.0761s; samplesPerSecond = 3600.0
MPI Rank 1: 12/12/2017 15:18:14:  Epoch[ 2 of 5]-Minibatch[  67-  69, 86.25%]: CrossEntropyWithSoftmax = 1.86749008 * 258; EvalClassificationError = 0.50775194 * 258; time = 0.0694s; samplesPerSecond = 3716.8
MPI Rank 1: 		(model aggregation stats): 3-th sync point was hit, introducing a 0.04-seconds latency this time; accumulated time on sync point = 0.04 seconds , average latency = 0.01 seconds
MPI Rank 1: 		(model aggregation stats) 3-th sync:     1.13 seconds since last report (0.45 seconds on comm.); 4246 samples processed by 2 workers (2102 by me);
MPI Rank 1: 		(model aggregation stats) 3-th sync: totalThroughput = 3.75k samplesPerSecond , throughputPerWorker = 1.87k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  70-  72, 90.00%]: CrossEntropyWithSoftmax = 1.98168893 * 271; EvalClassificationError = 0.56457565 * 271; time = 0.6235s; samplesPerSecond = 434.7
MPI Rank 1: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  73-  75, 93.75%]: CrossEntropyWithSoftmax = 2.01285094 * 278; EvalClassificationError = 0.53597122 * 278; time = 0.0767s; samplesPerSecond = 3623.7
MPI Rank 1: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  76-  78, 97.50%]: CrossEntropyWithSoftmax = 1.94638799 * 286; EvalClassificationError = 0.54895105 * 286; time = 0.0812s; samplesPerSecond = 3523.7
MPI Rank 1: 12/12/2017 15:18:15:  Epoch[ 2 of 5]-Minibatch[  79-  81, 101.25%]: CrossEntropyWithSoftmax = 1.95391112 * 170; EvalClassificationError = 0.53529412 * 170; time = 0.0439s; samplesPerSecond = 3870.4
MPI Rank 1: 		(model aggregation stats): 4-th sync point was hit, introducing a 0.29-seconds latency this time; accumulated time on sync point = 0.33 seconds , average latency = 0.08 seconds
MPI Rank 1: 		(model aggregation stats) 4-th sync:     2.07 seconds since last report (1.44 seconds on comm.); 7692 samples processed by 2 workers (904 by me);
MPI Rank 1: 		(model aggregation stats) 4-th sync: totalThroughput = 3.71k samplesPerSecond , throughputPerWorker = 1.86k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:17: Finished Epoch[ 2 of 5]: [Training] CrossEntropyWithSoftmax = 2.05813627 * 20480; EvalClassificationError = 0.56054688 * 20480; totalSamplesSeen = 40960; learningRatePerSample = 0.001953125; epochTime=5.61364s
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:17: Starting Epoch 3: learning rate per sample = 0.000098  effective momentum = 0.656119  momentum as time constant = 2429.9 samples
MPI Rank 1: Parallel training (2 workers) using BlockMomentumSGD with block momentum = 0.5000, block momentum time constant (per worker) = 2954.6394, block learning rate = 1.0000, block size per worker = 2048 samples, using Nesterov-style block momentum, resetting SGD momentum after sync.
MPI Rank 1: minibatchiterator: epoch 2: frames [40960..61440] (first utterance at frame 40960), data subset 1 of 2, with 1 datapasses
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:17: Starting minibatch loop, distributed reading is ENABLED.
MPI Rank 1: 12/12/2017 15:18:17:  Epoch[ 3 of 5]-Minibatch[   1-   3, 15.00%]: CrossEntropyWithSoftmax = 1.95258108 * 1130; EvalClassificationError = 0.53539823 * 1130; time = 0.3490s; samplesPerSecond = 3238.2
MPI Rank 1: 		(model aggregation stats): 1-th sync point was hit, introducing a 0.05-seconds latency this time; accumulated time on sync point = 0.05 seconds , average latency = 0.05 seconds
MPI Rank 1: 		(model aggregation stats) 1-th sync:     1.34 seconds since last report (0.34 seconds on comm.); 4885 samples processed by 2 workers (2293 by me);
MPI Rank 1: 		(model aggregation stats) 1-th sync: totalThroughput = 3.64k samplesPerSecond , throughputPerWorker = 1.82k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:18:  Epoch[ 3 of 5]-Minibatch[   4-   6, 30.00%]: CrossEntropyWithSoftmax = 1.88618176 * 1163; EvalClassificationError = 0.53224420 * 1163; time = 0.9916s; samplesPerSecond = 1172.8
MPI Rank 1: 12/12/2017 15:18:19:  Epoch[ 3 of 5]-Minibatch[   7-   9, 45.00%]: CrossEntropyWithSoftmax = 1.99391764 * 1085; EvalClassificationError = 0.54101382 * 1085; time = 0.2519s; samplesPerSecond = 4306.5
MPI Rank 1: 		(model aggregation stats): 2-th sync point was hit, introducing a 0.09-seconds latency this time; accumulated time on sync point = 0.14 seconds , average latency = 0.07 seconds
MPI Rank 1: 		(model aggregation stats) 2-th sync:     1.09 seconds since last report (0.40 seconds on comm.); 4826 samples processed by 2 workers (2249 by me);
MPI Rank 1: 		(model aggregation stats) 2-th sync: totalThroughput = 4.42k samplesPerSecond , throughputPerWorker = 2.21k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:20:  Epoch[ 3 of 5]-Minibatch[  10-  12, 60.00%]: CrossEntropyWithSoftmax = 1.94066838 * 1164; EvalClassificationError = 0.56701031 * 1164; time = 0.8402s; samplesPerSecond = 1385.4
MPI Rank 1: 12/12/2017 15:18:20:  Epoch[ 3 of 5]-Minibatch[  13-  15, 75.00%]: CrossEntropyWithSoftmax = 2.01263165 * 1167; EvalClassificationError = 0.57497858 * 1167; time = 0.2647s; samplesPerSecond = 4408.9
MPI Rank 1: 		(model aggregation stats): 3-th sync point was hit, introducing a 0.07-seconds latency this time; accumulated time on sync point = 0.21 seconds , average latency = 0.07 seconds
MPI Rank 1: 		(model aggregation stats) 3-th sync:     0.99 seconds since last report (0.13 seconds on comm.); 4903 samples processed by 2 workers (2326 by me);
MPI Rank 1: 		(model aggregation stats) 3-th sync: totalThroughput = 4.94k samplesPerSecond , throughputPerWorker = 2.47k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:21:  Epoch[ 3 of 5]-Minibatch[  16-  18, 90.00%]: CrossEntropyWithSoftmax = 1.93952077 * 1159; EvalClassificationError = 0.53321829 * 1159; time = 0.7278s; samplesPerSecond = 1592.5
MPI Rank 1: 12/12/2017 15:18:21:  Epoch[ 3 of 5]-Minibatch[  19-  21, 105.00%]: CrossEntropyWithSoftmax = 1.99579390 * 823; EvalClassificationError = 0.55285541 * 823; time = 0.2106s; samplesPerSecond = 3908.2
MPI Rank 1: 		(model aggregation stats): 4-th sync point was hit, introducing a 0.39-seconds latency this time; accumulated time on sync point = 0.60 seconds , average latency = 0.15 seconds
MPI Rank 1: 		(model aggregation stats) 4-th sync:     1.83 seconds since last report (1.00 seconds on comm.); 5866 samples processed by 2 workers (823 by me);
MPI Rank 1: 		(model aggregation stats) 4-th sync: totalThroughput = 3.21k samplesPerSecond , throughputPerWorker = 1.60k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:22: Finished Epoch[ 3 of 5]: [Training] CrossEntropyWithSoftmax = 1.96686676 * 20480; EvalClassificationError = 0.54916992 * 20480; totalSamplesSeen = 61440; learningRatePerSample = 9.7656251e-05; epochTime=5.312s
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:23: Starting Epoch 4: learning rate per sample = 0.000098  effective momentum = 0.656119  momentum as time constant = 2429.9 samples
MPI Rank 1: Parallel training (2 workers) using BlockMomentumSGD with block momentum = 0.5000, block momentum time constant (per worker) = 2954.6394, block learning rate = 1.0000, block size per worker = 2048 samples, using Nesterov-style block momentum, resetting SGD momentum after sync.
MPI Rank 1: minibatchiterator: epoch 3: frames [61440..81920] (first utterance at frame 61440), data subset 1 of 2, with 1 datapasses
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:23: Starting minibatch loop, distributed reading is ENABLED.
MPI Rank 1: 12/12/2017 15:18:23:  Epoch[ 4 of 5]-Minibatch[   1-   3, 15.00%]: CrossEntropyWithSoftmax = 1.90803387 * 1149; EvalClassificationError = 0.53176675 * 1149; time = 0.3063s; samplesPerSecond = 3751.7
MPI Rank 1: 		(model aggregation stats): 1-th sync point was hit, introducing a 0.03-seconds latency this time; accumulated time on sync point = 0.03 seconds , average latency = 0.03 seconds
MPI Rank 1: 		(model aggregation stats) 1-th sync:     0.92 seconds since last report (0.23 seconds on comm.); 4901 samples processed by 2 workers (2351 by me);
MPI Rank 1: 		(model aggregation stats) 1-th sync: totalThroughput = 5.33k samplesPerSecond , throughputPerWorker = 2.66k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:24:  Epoch[ 4 of 5]-Minibatch[   4-   6, 30.00%]: CrossEntropyWithSoftmax = 1.95574649 * 1202; EvalClassificationError = 0.53161398 * 1202; time = 0.6103s; samplesPerSecond = 1969.5
MPI Rank 1: 12/12/2017 15:18:24:  Epoch[ 4 of 5]-Minibatch[   7-   9, 45.00%]: CrossEntropyWithSoftmax = 1.91697276 * 1130; EvalClassificationError = 0.53362832 * 1130; time = 0.2557s; samplesPerSecond = 4419.2
MPI Rank 1: 		(model aggregation stats): 2-th sync point was hit, introducing a 0.04-seconds latency this time; accumulated time on sync point = 0.06 seconds , average latency = 0.03 seconds
MPI Rank 1: 		(model aggregation stats) 2-th sync:     1.01 seconds since last report (0.21 seconds on comm.); 4836 samples processed by 2 workers (2317 by me);
MPI Rank 1: 		(model aggregation stats) 2-th sync: totalThroughput = 4.81k samplesPerSecond , throughputPerWorker = 2.40k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:25:  Epoch[ 4 of 5]-Minibatch[  10-  12, 60.00%]: CrossEntropyWithSoftmax = 1.86120693 * 1187; EvalClassificationError = 0.51053075 * 1187; time = 0.7498s; samplesPerSecond = 1583.0
MPI Rank 1: 12/12/2017 15:18:25:  Epoch[ 4 of 5]-Minibatch[  13-  15, 75.00%]: CrossEntropyWithSoftmax = 1.95323579 * 1202; EvalClassificationError = 0.54991681 * 1202; time = 0.2753s; samplesPerSecond = 4366.6
MPI Rank 1: 		(model aggregation stats): 3-th sync point was hit, introducing a 0.07-seconds latency this time; accumulated time on sync point = 0.13 seconds , average latency = 0.04 seconds
MPI Rank 1: 		(model aggregation stats) 3-th sync:     0.86 seconds since last report (0.12 seconds on comm.); 4952 samples processed by 2 workers (2401 by me);
MPI Rank 1: 		(model aggregation stats) 3-th sync: totalThroughput = 5.73k samplesPerSecond , throughputPerWorker = 2.87k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:25:  Epoch[ 4 of 5]-Minibatch[  16-  18, 90.00%]: CrossEntropyWithSoftmax = 1.93023053 * 1199; EvalClassificationError = 0.53961635 * 1199; time = 0.5883s; samplesPerSecond = 2038.1
MPI Rank 1: 12/12/2017 15:18:26:  Epoch[ 4 of 5]-Minibatch[  19-  21, 105.00%]: CrossEntropyWithSoftmax = 1.93100189 * 817; EvalClassificationError = 0.54834761 * 817; time = 0.1816s; samplesPerSecond = 4500.0
MPI Rank 1: 		(model aggregation stats): 4-th sync point was hit, introducing a 0.37-seconds latency this time; accumulated time on sync point = 0.50 seconds , average latency = 0.12 seconds
MPI Rank 1: 		(model aggregation stats) 4-th sync:     1.51 seconds since last report (0.87 seconds on comm.); 5791 samples processed by 2 workers (817 by me);
MPI Rank 1: 		(model aggregation stats) 4-th sync: totalThroughput = 3.83k samplesPerSecond , throughputPerWorker = 1.92k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:27: Finished Epoch[ 4 of 5]: [Training] CrossEntropyWithSoftmax = 1.91580675 * 20480; EvalClassificationError = 0.53129883 * 20480; totalSamplesSeen = 81920; learningRatePerSample = 9.7656251e-05; epochTime=4.33567s
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:27: Starting Epoch 5: learning rate per sample = 0.000098  effective momentum = 0.656119  momentum as time constant = 2429.9 samples
MPI Rank 1: Parallel training (2 workers) using BlockMomentumSGD with block momentum = 0.5000, block momentum time constant (per worker) = 2954.6394, block learning rate = 1.0000, block size per worker = 2048 samples, using Nesterov-style block momentum, resetting SGD momentum after sync.
MPI Rank 1: minibatchiterator: epoch 4: frames [81920..102400] (first utterance at frame 81920), data subset 1 of 2, with 1 datapasses
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:27: Starting minibatch loop, distributed reading is ENABLED.
MPI Rank 1: 12/12/2017 15:18:28:  Epoch[ 5 of 5]-Minibatch[   1-   3, 15.00%]: CrossEntropyWithSoftmax = 1.88109892 * 1209; EvalClassificationError = 0.50124069 * 1209; time = 0.3167s; samplesPerSecond = 3817.6
MPI Rank 1: 		(model aggregation stats): 1-th sync point was hit, introducing a 0.00-seconds latency this time; accumulated time on sync point = 0.00 seconds , average latency = 0.00 seconds
MPI Rank 1: 		(model aggregation stats) 1-th sync:     1.05 seconds since last report (0.33 seconds on comm.); 4919 samples processed by 2 workers (2426 by me);
MPI Rank 1: 		(model aggregation stats) 1-th sync: totalThroughput = 4.68k samplesPerSecond , throughputPerWorker = 2.34k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:28:  Epoch[ 5 of 5]-Minibatch[   4-   6, 30.00%]: CrossEntropyWithSoftmax = 1.95030921 * 1217; EvalClassificationError = 0.53656532 * 1217; time = 0.7307s; samplesPerSecond = 1665.6
MPI Rank 1: 12/12/2017 15:18:29:  Epoch[ 5 of 5]-Minibatch[   7-   9, 45.00%]: CrossEntropyWithSoftmax = 1.90838044 * 1206; EvalClassificationError = 0.52155887 * 1206; time = 0.2824s; samplesPerSecond = 4270.1
MPI Rank 1: 		(model aggregation stats): 2-th sync point was hit, introducing a 0.02-seconds latency this time; accumulated time on sync point = 0.02 seconds , average latency = 0.01 seconds
MPI Rank 1: 		(model aggregation stats) 2-th sync:     1.23 seconds since last report (0.31 seconds on comm.); 4899 samples processed by 2 workers (2419 by me);
MPI Rank 1: 		(model aggregation stats) 2-th sync: totalThroughput = 3.99k samplesPerSecond , throughputPerWorker = 1.99k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:29:  Epoch[ 5 of 5]-Minibatch[  10-  12, 60.00%]: CrossEntropyWithSoftmax = 1.97427625 * 1213; EvalClassificationError = 0.54492993 * 1213; time = 0.9459s; samplesPerSecond = 1282.3
MPI Rank 1: 12/12/2017 15:18:30:  Epoch[ 5 of 5]-Minibatch[  13-  15, 75.00%]: CrossEntropyWithSoftmax = 1.85531359 * 1147; EvalClassificationError = 0.51612903 * 1147; time = 0.2739s; samplesPerSecond = 4187.5
MPI Rank 1: 		(model aggregation stats): 3-th sync point was hit, introducing a 0.08-seconds latency this time; accumulated time on sync point = 0.10 seconds , average latency = 0.03 seconds
MPI Rank 1: 		(model aggregation stats) 3-th sync:     0.85 seconds since last report (0.14 seconds on comm.); 4829 samples processed by 2 workers (2359 by me);
MPI Rank 1: 		(model aggregation stats) 3-th sync: totalThroughput = 5.71k samplesPerSecond , throughputPerWorker = 2.85k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:30:  Epoch[ 5 of 5]-Minibatch[  16-  18, 90.00%]: CrossEntropyWithSoftmax = 2.03114739 * 1212; EvalClassificationError = 0.56023102 * 1212; time = 0.5720s; samplesPerSecond = 2118.9
MPI Rank 1: 12/12/2017 15:18:31:  Epoch[ 5 of 5]-Minibatch[  19-  21, 105.00%]: CrossEntropyWithSoftmax = 1.89611181 * 809; EvalClassificationError = 0.51174289 * 809; time = 0.2218s; samplesPerSecond = 3647.3
MPI Rank 1: 		(model aggregation stats): 4-th sync point was hit, introducing a 0.82-seconds latency this time; accumulated time on sync point = 0.92 seconds , average latency = 0.23 seconds
MPI Rank 1: 		(model aggregation stats) 4-th sync:     1.91 seconds since last report (0.74 seconds on comm.); 5833 samples processed by 2 workers (809 by me);
MPI Rank 1: 		(model aggregation stats) 4-th sync: totalThroughput = 3.06k samplesPerSecond , throughputPerWorker = 1.53k samplesPerSecond
MPI Rank 1: 12/12/2017 15:18:32: Finished Epoch[ 5 of 5]: [Training] CrossEntropyWithSoftmax = 1.89863387 * 20480; EvalClassificationError = 0.52202148 * 20480; totalSamplesSeen = 102400; learningRatePerSample = 9.7656251e-05; epochTime=5.12581s
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:32: Action "train" complete.
MPI Rank 1: 
MPI Rank 1: 12/12/2017 15:18:32: __COMPLETED__