CPU info:
    CPU Model Name: Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz
    Hardware threads: 12
    Total Memory: 33474872 kB
-------------------------------------------------------------------
++ python -c 'import sys; sys.stdout.write(str(sys.version_info[0])); sys.stdout.flush()'
+ PythonVer=2
+ '[' 2 == 3 ']'
+ [[ C:\CNTKTestData == '' ]]
+ [[ ! -d C:\CNTKTestData ]]
+ '[' Windows_NT == Windows_NT ']'
++ cygpath -au 'C:\CNTKTestData'
+ TestDataSourceDir=/cygdrive/c/CNTKTestData
+ '[' cpu == cpu ']'
+ export CUDA_VISIBLE_DEVICES=-1
+ CUDA_VISIBLE_DEVICES=-1
+ '[' cpu == cpu ']'
+ echo Use pre-trained resnet_20_cifar_10 model and atis.dnn in 'C:\CNTKTestData' for tests on CPU device.
Use pre-trained resnet_20_cifar_10 model and atis.dnn in C:\CNTKTestData for tests on CPU device.
+ python train_models_for_evaluation.py -u -d cpu
the output_dir is /cygdrive/c/repos/ThirdCNTK/Tests/EndToEndTests/EvalClientTests/CNTKLibraryCSEvalExamplesTest.
the test_device is cpu.
unzip_only is True
+ cp /cygdrive/c/CNTKTestData/PreTrainedModels/EvalModels/resnet20_cifar10_python.dnn resnet20.dnn
+ cp /cygdrive/c/CNTKTestData/PreTrainedModels/EvalModels/atis.dnn atis.dnn
+ mv data/train/00000.png data/train/00001.png data/train/00002.png data/train/00003.png data/train/00004.png data/train/00005.png .
+ cp /cygdrive/c/repos/ThirdCNTK/Tests/EndToEndTests/../../Examples/LanguageUnderstanding/ATIS/BrainScript/query.wl .
+ cp /cygdrive/c/repos/ThirdCNTK/Tests/EndToEndTests/../../Examples/LanguageUnderstanding/ATIS/BrainScript/slots.wl .
+ '[' Windows_NT == Windows_NT ']'
+ /cygdrive/c/repos/ThirdCNTK/x64/release/CNTKLibraryCSEvalExamplesTest.exe
Test Utils
MaxNumCPUThreads: before: 12, after 2
reset MaxNumCPuThreads to 12
TraceLevel: before: Warning, after Info
reset TraceLevel to Warning
Float
8
CPU
GPU
False
True
True
Constant
Placeholder
Input
Output
Parameter
======== Evaluate model on CPU using CPUOnly build ========

===== Evaluate single image =====
Evaluation result for image 00000.png
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057733,-1.921682,6.629186,-3.328488,1.398772,5.083784,-2.13107,2.858712,-3.295478.

===== Setup memory tests using Resnet Model =====

Test object reference inside SetupUsingResetModel.


Print out saved object references.
Device0: CPU, Type: CPU
Axis0: defaultBatchAxis, IsStaticAxis: False
OutputVar: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
OutputVar0: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
InputVar0: Parameter('W', [10 x 1 x 1 x 64], []), Name: W, Kind: Output, Shape: [10 x 1 x 1 x 64]
ArgumentVar0: Input('features', [32 x 32 x 3], [#]), Name: features, Kind: Output, Shape: [32 x 32 x 3]
OutputVal: , Device: CPU, Storage: Dense, Shape: [10 x 2]Data:
The number of sequences in the batch: 2
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057731,-1.921683,6.629186,-3.32849,1.398771,5.083781,-2.131069,2.858713,-3.295477.
Sequence 1 contains 1 samples.
    sample 0: -2.999754,1.988537,-6.157244,-0.8313794,-6.47598,-0.4141227,-3.218865,-1.733618,5.188062,14.60122.

All saved object references are printed.
Test creating NDArrayView on devices.
n1: on CPU, Storage:Dense, Shape:[10]
n1Clone: on CPU, Storage:Dense, Shape:[10]
n1CloneCPU: on CPU, Storage:Dense, Shape:[10]
s1: on CPU, Storage:SparseCSC, Shape:[4 x 5]
s1Clone: on CPU, Storage:SparseCSC, Shape:[4 x 5]
s1DenseCPU: on CPU, Storage:Dense, Shape:[4 x 5]

1. Run: Test saved object references.


Print out saved object references.
Device0: CPU, Type: CPU
Axis0: defaultBatchAxis, IsStaticAxis: False
OutputVar: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
OutputVar0: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
InputVar0: Parameter('W', [10 x 1 x 1 x 64], []), Name: W, Kind: Output, Shape: [10 x 1 x 1 x 64]
ArgumentVar0: Input('features', [32 x 32 x 3], [#]), Name: features, Kind: Output, Shape: [32 x 32 x 3]
OutputVal: , Device: CPU, Storage: Dense, Shape: [10 x 2]Data:
The number of sequences in the batch: 2
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057731,-1.921683,6.629186,-3.32849,1.398771,5.083781,-2.131069,2.858713,-3.295477.
Sequence 1 contains 1 samples.
    sample 0: -2.999754,1.988537,-6.157244,-0.8313794,-6.47598,-0.4141227,-3.218865,-1.733618,5.188062,14.60122.

All saved object references are printed.

2. Run: Test saved object references.


Print out saved object references.
Device0: CPU, Type: CPU
Axis0: defaultBatchAxis, IsStaticAxis: False
OutputVar: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
OutputVar0: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
InputVar0: Parameter('W', [10 x 1 x 1 x 64], []), Name: W, Kind: Output, Shape: [10 x 1 x 1 x 64]
ArgumentVar0: Input('features', [32 x 32 x 3], [#]), Name: features, Kind: Output, Shape: [32 x 32 x 3]
OutputVal: , Device: CPU, Storage: Dense, Shape: [10 x 2]Data:
The number of sequences in the batch: 2
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057731,-1.921683,6.629186,-3.32849,1.398771,5.083781,-2.131069,2.858713,-3.295477.
Sequence 1 contains 1 samples.
    sample 0: -2.999754,1.988537,-6.157244,-0.8313794,-6.47598,-0.4141227,-3.218865,-1.733618,5.188062,14.60122.

All saved object references are printed.

===== Evaluate batch of images =====
Evaluation result for batch of images: 00000.png, 00001.png, 00002.png
The number of sequences in the batch: 3
Sequence 0 contains 1 samples.
    sample 0: -2.197929,-3.057732,-1.921682,6.629186,-3.328489,1.398772,5.083784,-2.131071,2.858711,-3.295479.
Sequence 1 contains 1 samples.
    sample 0: -2.999755,1.988538,-6.157244,-0.8313787,-6.475982,-0.4141223,-3.218866,-1.733616,5.188062,14.60122.
Sequence 2 contains 1 samples.
    sample 0: 6.493433,-3.925384,1.195913,1.00218,-1.663995,-4.729935,-1.476986,-0.4302735,0.115595,3.42197.

Print out saved object references.
Device0: CPU, Type: CPU
Axis0: defaultBatchAxis, IsStaticAxis: False
OutputVar: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
OutputVar0: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
InputVar0: Parameter('W', [10 x 1 x 1 x 64], []), Name: W, Kind: Output, Shape: [10 x 1 x 1 x 64]
ArgumentVar0: Input('features', [32 x 32 x 3], [#]), Name: features, Kind: Output, Shape: [32 x 32 x 3]
OutputVal: , Device: CPU, Storage: Dense, Shape: [10 x 2]Data:
The number of sequences in the batch: 2
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057731,-1.921683,6.629186,-3.32849,1.398771,5.083781,-2.131069,2.858713,-3.295477.
Sequence 1 contains 1 samples.
    sample 0: -2.999754,1.988537,-6.157244,-0.8313794,-6.47598,-0.4141227,-3.218865,-1.733618,5.188062,14.60122.

All saved object references are printed.

===== Evaluate multiple images in parallel =====
Evaluate 5 images in parallel using 3 model instances.
Evaluating image 00000.png using thread 9.
Evaluating image 00001.png using thread 7.
Evaluating image 00002.png using thread 16.
Evaluating image 00003.png using thread 9.
Evaluating image 00004.png using thread 7.
Evaluation result for image 00000.png
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057733,-1.921682,6.629186,-3.328488,1.398772,5.083784,-2.13107,2.858712,-3.295478.
Evaluation result for image 00001.png
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: -2.999755,1.988537,-6.157243,-0.8313792,-6.475981,-0.4141223,-3.218865,-1.733617,5.188062,14.60122.
Evaluation result for image 00002.png
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: 6.493433,-3.925385,1.195913,1.00218,-1.663993,-4.729934,-1.476986,-0.4302738,0.1155958,3.421968.
Evaluation result for image 00003.png
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: -1.07106,-6.325305,-1.910145,-0.05163154,10.90496,0.5978318,-1.849957,1.134093,0.3312155,-1.688351.
Evaluation result for image 00004.png
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: 0.0114498,18.66314,-4.937451,-1.941423,-8.815382,-0.4959869,-1.087446,-3.517007,-2.210147,4.337873.

===== Setup memory tests using Resnet Model =====

Test object reference inside SetupUsingResetModel.


Print out saved object references.
Device0: CPU, Type: CPU
Axis0: defaultBatchAxis, IsStaticAxis: False
OutputVar: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
OutputVar0: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
InputVar0: Parameter('W', [10 x 1 x 1 x 64], []), Name: W, Kind: Output, Shape: [10 x 1 x 1 x 64]
ArgumentVar0: Input('features', [32 x 32 x 3], [#]), Name: features, Kind: Output, Shape: [32 x 32 x 3]
OutputVal: , Device: CPU, Storage: Dense, Shape: [10 x 2]Data:
The number of sequences in the batch: 2
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057731,-1.921683,6.629186,-3.32849,1.398771,5.083781,-2.131069,2.858713,-3.295477.
Sequence 1 contains 1 samples.
    sample 0: -2.999754,1.988537,-6.157244,-0.8313794,-6.47598,-0.4141227,-3.218865,-1.733618,5.188062,14.60122.

All saved object references are printed.
Test creating NDArrayView on devices.
n1: on CPU, Storage:Dense, Shape:[10]
n1Clone: on CPU, Storage:Dense, Shape:[10]
n1CloneCPU: on CPU, Storage:Dense, Shape:[10]
s1: on CPU, Storage:SparseCSC, Shape:[4 x 5]
s1Clone: on CPU, Storage:SparseCSC, Shape:[4 x 5]
s1DenseCPU: on CPU, Storage:Dense, Shape:[4 x 5]

1. Run: Test saved object references.


Print out saved object references.
Device0: CPU, Type: CPU
Axis0: defaultBatchAxis, IsStaticAxis: False
OutputVar: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
OutputVar0: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
InputVar0: Parameter('W', [10 x 1 x 1 x 64], []), Name: W, Kind: Output, Shape: [10 x 1 x 1 x 64]
ArgumentVar0: Input('features', [32 x 32 x 3], [#]), Name: features, Kind: Output, Shape: [32 x 32 x 3]
OutputVal: , Device: CPU, Storage: Dense, Shape: [10 x 2]Data:
The number of sequences in the batch: 2
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057731,-1.921683,6.629186,-3.32849,1.398771,5.083781,-2.131069,2.858713,-3.295477.
Sequence 1 contains 1 samples.
    sample 0: -2.999754,1.988537,-6.157244,-0.8313794,-6.47598,-0.4141227,-3.218865,-1.733618,5.188062,14.60122.

All saved object references are printed.

2. Run: Test saved object references.


Print out saved object references.
Device0: CPU, Type: CPU
Axis0: defaultBatchAxis, IsStaticAxis: False
OutputVar: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
OutputVar0: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
InputVar0: Parameter('W', [10 x 1 x 1 x 64], []), Name: W, Kind: Output, Shape: [10 x 1 x 1 x 64]
ArgumentVar0: Input('features', [32 x 32 x 3], [#]), Name: features, Kind: Output, Shape: [32 x 32 x 3]
OutputVal: , Device: CPU, Storage: Dense, Shape: [10 x 2]Data:
The number of sequences in the batch: 2
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057731,-1.921683,6.629186,-3.32849,1.398771,5.083781,-2.131069,2.858713,-3.295477.
Sequence 1 contains 1 samples.
    sample 0: -2.999754,1.988537,-6.157244,-0.8313794,-6.47598,-0.4141227,-3.218865,-1.733618,5.188062,14.60122.

All saved object references are printed.

===== Evaluate image asynchronously =====
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057733,-1.921682,6.629186,-3.328488,1.398772,5.083784,-2.13107,2.858712,-3.295478.

===== Evaluate single sequence using one-hot vector =====
            BOS ---- O
              i ---- O
          would ---- O
           like ---- O
             to ---- O
           find ---- O
              a ---- O
         flight ---- O
           from ---- O
      charlotte ---- B-fromloc.city_name
             to ---- O
            las ---- B-toloc.city_name
          vegas ---- I-toloc.city_name
           that ---- O
          makes ---- O
              a ---- O
           stop ---- O
             in ---- O
            st. ---- B-stoploc.city_name
          louis ---- B-arrive_time.time_relative
            EOS ---- O


===== Evaluate batch of sequences with variable length using one-hot vector =====
The number of sequences in the batch: 2
Sequence 0: 
            BOS ---- O
              i ---- O
          would ---- O
           like ---- O
             to ---- O
           find ---- O
              a ---- O
         flight ---- O
           from ---- O
      charlotte ---- B-fromloc.city_name
             to ---- O
            las ---- B-toloc.city_name
          vegas ---- I-toloc.city_name
           that ---- O
          makes ---- O
              a ---- O
           stop ---- O
             in ---- O
            st. ---- B-stoploc.city_name
          louis ---- B-arrive_time.time_relative
            EOS ---- O

Sequence 1: 
            BOS ---- O
        flights ---- O
           from ---- O
            new ---- B-fromloc.city_name
           york ---- I-fromloc.city_name
             to ---- O
        seattle ---- B-toloc.city_name
            EOS ---- O


===== Evaluate single sequence using sparse input =====
            BOS ---- O
              i ---- O
          would ---- O
           like ---- O
             to ---- O
           find ---- O
              a ---- O
         flight ---- O
           from ---- O
      charlotte ---- B-fromloc.city_name
             to ---- O
            las ---- B-toloc.city_name
          vegas ---- I-toloc.city_name
           that ---- O
          makes ---- O
              a ---- O
           stop ---- O
             in ---- O
            st. ---- B-stoploc.city_name
          louis ---- B-arrive_time.time_relative
            EOS ---- O


===== Load model from memory buffer =====
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057733,-1.921682,6.629186,-3.328488,1.398772,5.083784,-2.13107,2.858712,-3.295478.

Print out saved object references.
Device0: CPU, Type: CPU
Axis0: defaultBatchAxis, IsStaticAxis: False
OutputVar: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
OutputVar0: Output('Block5480_Output_0', [10], [#]), Name: , Kind: Output, Shape: [10]
InputVar0: Parameter('W', [10 x 1 x 1 x 64], []), Name: W, Kind: Output, Shape: [10 x 1 x 1 x 64]
ArgumentVar0: Input('features', [32 x 32 x 3], [#]), Name: features, Kind: Output, Shape: [32 x 32 x 3]
OutputVal: , Device: CPU, Storage: Dense, Shape: [10 x 2]Data:
The number of sequences in the batch: 2
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057731,-1.921683,6.629186,-3.32849,1.398771,5.083781,-2.131069,2.858713,-3.295477.
Sequence 1 contains 1 samples.
    sample 0: -2.999754,1.988537,-6.157244,-0.8313794,-6.47598,-0.4141227,-3.218865,-1.733618,5.188062,14.60122.

All saved object references are printed.

===== Evaluate intermediate layer =====

Evaluation result of intermediate layer final_avg_pooling
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: 0.8178836,0.8373678,0.6518813,0.3843705,0.6911097,0.3901796,1.010936,0.09253094,0.5516857,1.308782,0.2308613,1.26787,0.6026684,0.7031508,1.110137,0.4074853,0.4242566,0.2809644,0.900517,0.4175932,0.3740564,0.08452793,0.1951823,0.7342843,0.7735909,0.1431717,0.367403,0.01266347,0.2634773,0.6433478,0.3033065,0.372538,0.4073464,1.434568,0.04759531,0.2470742,0.1162971,0.731542,0.3109912,0.193792,0.4229452,0.1753684,1.201418,0.6783847,0.4534732,0.05665915,0.1267008,0.2703206,0.04396501,0.07115713,0.2515253,0.1316103,0.2947579,0.2722122,0.01376147,0.796611,0.6232213,0.08000247,0.06987222,0.08726446,0.01344907,0.7560263,0.3477854,0.01986465.

===== Evaluate combined outputs =====

Evaluation result of last layer of the model
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: -2.19793,-3.057733,-1.921682,6.629186,-3.328488,1.398772,5.083784,-2.13107,2.858712,-3.295478.
Evaluation result of intermediate layer final_avg_pooling
The number of sequences in the batch: 1
Sequence 0 contains 1 samples.
    sample 0: 0.8178836,0.8373678,0.6518813,0.3843705,0.6911097,0.3901796,1.010936,0.09253094,0.5516857,1.308782,0.2308613,1.26787,0.6026684,0.7031508,1.110137,0.4074853,0.4242566,0.2809644,0.900517,0.4175932,0.3740564,0.08452793,0.1951823,0.7342843,0.7735909,0.1431717,0.367403,0.01266347,0.2634773,0.6433478,0.3033065,0.372538,0.4073464,1.434568,0.04759531,0.2470742,0.1162971,0.731542,0.3109912,0.193792,0.4229452,0.1753684,1.201418,0.6783847,0.4534732,0.05665915,0.1267008,0.2703206,0.04396501,0.07115713,0.2515253,0.1316103,0.2947579,0.2722122,0.01376147,0.796611,0.6232213,0.08000247,0.06987222,0.08726446,0.01344907,0.7560263,0.3477854,0.01986465.
======== Evaluation completes. ========
+ ExitCode=0
+ exit 0