Merge branch 'qiwye/lightrnn' of https://github.com/Microsoft/CNTK in…

…to qiwye/lightrnn

Conflicts:
	Tests/EndToEndTests/Downloads/run-test

.gitattributes

@@ -144,6 +144,12 @@ Examples/Text/LightRNN/test/word-0.location text
 *.dnn binary
 Examples/Image/Detection/FastRCNN/fastRCNN/*/*.pyd binary
 Examples/Image/Detection/FastRCNN/fastRCNN/*/*.so binary
+Examples/Image/Detection/utils/cython_modules/*.pyd binary
+Examples/Image/Detection/utils/cython_modules/*.so binary
 Tests/UnitTests/V2LibraryTests/data/*.bin binary
 Tests/UnitTests/ReaderTests/Data/CNTKBinaryReader/*.bin binary
 Tests/EndToEndTests/ParallelTraining/AsynchronousSGD/ASGD_Resnet.model.1 binary
+Examples/Extensibility/BinaryConvolution/BinaryConvolutionLib/halide/halide_convolve.a binary
+Examples/Extensibility/BinaryConvolution/BinaryConvolutionLib/halide/halide_convolve.lib binary
+Examples/Extensibility/BinaryConvolution/BinaryConvolutionLib/halide/halide_convolve_nofeatures.a binary
+Examples/Extensibility/BinaryConvolution/BinaryConvolutionLib/halide/halide_convolve_nofeatures.lib binary

.gitignore

@@ -8,6 +8,7 @@
 *.orig
 \#*
 .\#*
+.vscode
 
 # Local build configuration
 Config.make
@@ -292,6 +293,9 @@ Examples/Video/DataSets/UCF11/action_youtube_naudio/
 Examples/Video/DataSets/UCF11/test_map.csv
 Examples/Video/DataSets/UCF11/train_map.csv
 Tutorials/zip
+Manual/*.tmp
+Manual/*.ckp
+Manual/.ipynb_checkpoints
 
 # other
 /packages

CNTK.Common.props

@@ -29,7 +29,7 @@
     <HasJava>false</HasJava>
     <HasJava Condition="Exists('$(JAVA_HOME)\bin\javac.exe')">true</HasJava>
 
-    <CntkComponentVersion>2.0rc2</CntkComponentVersion>
+    <CntkComponentVersion>2.0</CntkComponentVersion>
     <CntkComponentVersion Condition="$(DebugBuild)">$(CntkComponentVersion)d</CntkComponentVersion>
   </PropertyGroup>
 

CNTK.Cpp.props

@@ -97,12 +97,27 @@
     <CudaPath>$(CUDA_PATH_V8_0)</CudaPath>
     <CudaRuntimeDll>cudart64_80.dll</CudaRuntimeDll>
     <CudaDlls>cublas64_80.dll;cusparse64_80.dll;curand64_80.dll;$(CudaRuntimeDll)</CudaDlls>
+
+<!-- Use NvidiaCompute to define nvcc target architectures (will generate code to support them all, i.e. fat-binary, in release mode)
+    In debug mode we only include cubin/PTX for 30 and rely on PTX / JIT to generate the required native cubin format
+    http://docs.nvidia.com/cuda/pascal-compatibility-guide/index.html#building-applications-with-pascal-support -->
+    <NvidiaCompute Condition="$(DebugBuild)">$(CNTK_CUDA_CODEGEN_DEBUG)</NvidiaCompute>
+    <NvidiaCompute Condition="$(DebugBuild) And '$(NvidiaCompute)'==''">compute_30,sm_30</NvidiaCompute>
+
+    <NvidiaCompute Condition="$(ReleaseBuild)">$(CNTK_CUDA_CODEGEN_RELEASE)</NvidiaCompute>
+    <NvidiaCompute Condition="$(ReleaseBuild) And '$(NvidiaCompute)'==''">compute_30,sm_30;compute_35,sm_35;compute_50,sm_50;compute_60,sm_60;compute_61,sm_61</NvidiaCompute>
   </PropertyGroup>
 
   <PropertyGroup Condition="'$(CudaVersion)' == '7.5'">
     <CudaPath>$(CUDA_PATH_V7_5)</CudaPath>
     <CudaRuntimeDll>cudart64_75.dll</CudaRuntimeDll>
     <CudaDlls>cublas64_75.dll;cusparse64_75.dll;curand64_75.dll;$(CudaRuntimeDll)</CudaDlls>
+
+    <NvidiaCompute Condition="$(DebugBuild)">$(CNTK_CUDA_CODEGEN_DEBUG)</NvidiaCompute>
+    <NvidiaCompute Condition="$(DebugBuild) And '$(NvidiaCompute)'==''">compute_30,sm_30</NvidiaCompute>
+
+    <NvidiaCompute Condition="$(ReleaseBuild)">$(CNTK_CUDA_CODEGEN_RELEASE)</NvidiaCompute>
+    <NvidiaCompute Condition="$(ReleaseBuild) And '$(NvidiaCompute)'==''">compute_30,sm_30;compute_35,sm_35;compute_50,sm_50</NvidiaCompute>
   </PropertyGroup>
 
   <PropertyGroup>

CNTK.sln

@@ -858,8 +858,12 @@ Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "CNTKv2LibraryDll", "Source\
 EndProject
 Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "V2LibraryTests", "Tests\UnitTests\V2LibraryTests\V2LibraryTests.vcxproj", "{F4CC3AB2-0DB2-4281-929A-2E68E30F0F6E}"
 	ProjectSection(ProjectDependencies) = postProject
+		{33D2FD22-DEF2-4507-A58A-368F641AEBE5} = {33D2FD22-DEF2-4507-A58A-368F641AEBE5}
+		{7B7A563D-AA8E-4660-A805-D50235A02120} = {7B7A563D-AA8E-4660-A805-D50235A02120}
 		{91973E60-A7BE-4C86-8FDB-59C88A0B3715} = {91973E60-A7BE-4C86-8FDB-59C88A0B3715}
+		{E6F26F9A-FF64-4F0A-B749-CD309EE357EE} = {E6F26F9A-FF64-4F0A-B749-CD309EE357EE}
 		{E5606ECE-48CA-4464-BB12-09D81D02B9EF} = {E5606ECE-48CA-4464-BB12-09D81D02B9EF}
+		{7B7A51ED-AA8E-4660-A805-D50235A02120} = {7B7A51ED-AA8E-4660-A805-D50235A02120}
 	EndProjectSection
 EndProject
 Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "Scripts", "Scripts", "{68263A2F-1D5F-4C46-B5AF-2304B80FC3D4}"
@@ -951,14 +955,15 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "IRMetric", "IRMetric", "{E8
 EndProject
 Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "PythonBindings", "bindings\python\PythonBindings.vcxproj", "{CD721536-CFD3-413E-A3D7-FB0FAF989635}"
 	ProjectSection(ProjectDependencies) = postProject
-		{1D5787D4-52E4-45DB-951B-82F220EE0C6A} = {1D5787D4-52E4-45DB-951B-82F220EE0C6A}
+		{9BD0A711-0BBD-45B6-B81C-053F03C26CFB} = {9BD0A711-0BBD-45B6-B81C-053F03C26CFB}
 		{4B442D34-641A-4B37-9A4B-D18DBE28A979} = {4B442D34-641A-4B37-9A4B-D18DBE28A979}
-		{4CF94A50-0D17-432A-8B5A-8458E91C44A6} = {4CF94A50-0D17-432A-8B5A-8458E91C44A6}
-		{60BDB847-D0C4-4FD3-A947-0C15C08BCDB5} = {60BDB847-D0C4-4FD3-A947-0C15C08BCDB5}
 		{7B7A563D-AA8E-4660-A805-D50235A02120} = {7B7A563D-AA8E-4660-A805-D50235A02120}
+		{60BDB847-D0C4-4FD3-A947-0C15C08BCDB5} = {60BDB847-D0C4-4FD3-A947-0C15C08BCDB5}
+		{20DEE94F-2802-40B1-B88B-22755A03AA48} = {20DEE94F-2802-40B1-B88B-22755A03AA48}
+		{4CF94A50-0D17-432A-8B5A-8458E91C44A6} = {4CF94A50-0D17-432A-8B5A-8458E91C44A6}
 		{91973E60-A7BE-4C86-8FDB-59C88A0B3715} = {91973E60-A7BE-4C86-8FDB-59C88A0B3715}
-		{9BD0A711-0BBD-45B6-B81C-053F03C26CFB} = {9BD0A711-0BBD-45B6-B81C-053F03C26CFB}
 		{E5606ECE-48CA-4464-BB12-09D81D02B9EF} = {E5606ECE-48CA-4464-BB12-09D81D02B9EF}
+		{1D5787D4-52E4-45DB-951B-82F220EE0C6A} = {1D5787D4-52E4-45DB-951B-82F220EE0C6A}
 	EndProjectSection
 EndProject
 Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "Multiverso", "Source\Multiverso\src\Multiverso.vcxproj", "{16F14058-B116-49D9-8BA0-209F3AFFE849}"
@@ -992,6 +997,7 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "SLUHandsOn", "SLUHandsOn",
 	ProjectSection(SolutionItems) = preProject
 		Tutorials\SLUHandsOn\atis.test.ctf = Tutorials\SLUHandsOn\atis.test.ctf
 		Tutorials\SLUHandsOn\atis.train.ctf = Tutorials\SLUHandsOn\atis.train.ctf
+		Tutorials\CNTK_050_GuidedTour.ipynb = Tutorials\CNTK_050_GuidedTour.ipynb
 		Tutorials\SLUHandsOn\README.md = Tutorials\SLUHandsOn\README.md
 		Tutorials\SLUHandsOn\slu.forward.backward.cmf = Tutorials\SLUHandsOn\slu.forward.backward.cmf
 		Tutorials\SLUHandsOn\slu.forward.cmf = Tutorials\SLUHandsOn\slu.forward.cmf
@@ -1128,7 +1134,6 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "Python", "Python", "{58902B
 	ProjectSection(SolutionItems) = preProject
 		Examples\Image\Classification\ResNet\Python\README.md = Examples\Image\Classification\ResNet\Python\README.md
 		Examples\Image\Classification\ResNet\Python\resnet_models.py = Examples\Image\Classification\ResNet\Python\resnet_models.py
-		Examples\Image\Classification\ResNet\Python\run_Distributed.py = Examples\Image\Classification\ResNet\Python\run_Distributed.py
 		Examples\Image\Classification\ResNet\Python\TrainResNet_CIFAR10.py = Examples\Image\Classification\ResNet\Python\TrainResNet_CIFAR10.py
 		Examples\Image\Classification\ResNet\Python\TrainResNet_CIFAR10_Distributed.py = Examples\Image\Classification\ResNet\Python\TrainResNet_CIFAR10_Distributed.py
 	EndProjectSection
@@ -1146,8 +1151,25 @@ Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "BrainScript", "BrainScript"
 EndProject
 Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "FastRCNN", "FastRCNN", "{4EAFF1B2-2D70-4486-B95E-684E39A50609}"
 	ProjectSection(SolutionItems) = preProject
+		Examples\Image\Detection\FastRCNN\__init__.py = Examples\Image\Detection\FastRCNN\__init__.py
+		Examples\Image\Detection\FastRCNN\A1_GenerateInputROIs.py = Examples\Image\Detection\FastRCNN\A1_GenerateInputROIs.py
+		Examples\Image\Detection\FastRCNN\A2_RunWithBSModel.py = Examples\Image\Detection\FastRCNN\A2_RunWithBSModel.py
+		Examples\Image\Detection\FastRCNN\A2_RunWithPyModel.py = Examples\Image\Detection\FastRCNN\A2_RunWithPyModel.py
+		Examples\Image\Detection\FastRCNN\A3_ParseAndEvaluateOutput.py = Examples\Image\Detection\FastRCNN\A3_ParseAndEvaluateOutput.py
+		Examples\Image\Detection\FastRCNN\B1_VisualizeInputROIs.py = Examples\Image\Detection\FastRCNN\B1_VisualizeInputROIs.py
+		Examples\Image\Detection\FastRCNN\B2_EvaluateInputROIs.py = Examples\Image\Detection\FastRCNN\B2_EvaluateInputROIs.py
+		Examples\Image\Detection\FastRCNN\B3_VisualizeOutputROIs.py = Examples\Image\Detection\FastRCNN\B3_VisualizeOutputROIs.py
+		Examples\Image\Detection\FastRCNN\C1_DrawBboxesOnImages.py = Examples\Image\Detection\FastRCNN\C1_DrawBboxesOnImages.py
+		Examples\Image\Detection\FastRCNN\C2_AssignLabelsToBboxes.py = Examples\Image\Detection\FastRCNN\C2_AssignLabelsToBboxes.py
+		Examples\Image\Detection\FastRCNN\CNTK_FastRCNN_Eval.ipynb = Examples\Image\Detection\FastRCNN\CNTK_FastRCNN_Eval.ipynb
+		Examples\Image\Detection\FastRCNN\cntk_helpers.py = Examples\Image\Detection\FastRCNN\cntk_helpers.py
 		Examples\Image\Detection\FastRCNN\fastrcnn.cntk = Examples\Image\Detection\FastRCNN\fastrcnn.cntk
+		Examples\Image\Detection\FastRCNN\imdb_data.py = Examples\Image\Detection\FastRCNN\imdb_data.py
+		Examples\Image\Detection\FastRCNN\install_fastrcnn.py = Examples\Image\Detection\FastRCNN\install_fastrcnn.py
+		Examples\Image\Detection\FastRCNN\LabelMeConverter.py = Examples\Image\Detection\FastRCNN\LabelMeConverter.py
+		Examples\Image\Detection\FastRCNN\PARAMETERS.py = Examples\Image\Detection\FastRCNN\PARAMETERS.py
 		Examples\Image\Detection\FastRCNN\README.md = Examples\Image\Detection\FastRCNN\README.md
+		Examples\Image\Detection\FastRCNN\requirements.txt = Examples\Image\Detection\FastRCNN\requirements.txt
 	EndProjectSection
 EndProject
 Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "LanguageUnderstanding", "LanguageUnderstanding", "{11FBC2F7-5340-4E19-9BDC-1140D54817E2}"
@@ -1536,6 +1558,32 @@ Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "CNTKLibraryJavaBinding", "b
 		{E5606ECE-48CA-4464-BB12-09D81D02B9EF} = {E5606ECE-48CA-4464-BB12-09D81D02B9EF}
 	EndProjectSection
 EndProject
+Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "BinaryConvolution", "BinaryConvolution", "{65649688-3377-4FA9-8CD0-BDC3AC72E0AD}"
+EndProject
+Project("{8BC9CEB8-8B4A-11D0-8D11-00A0C91BC942}") = "BinaryConvolutionLib", "Examples\Extensibility\BinaryConvolution\BinaryConvolutionLib\BinaryConvolutionLib.vcxproj", "{20DEE94F-2802-40B1-B88B-22755A03AA48}"
+	ProjectSection(ProjectDependencies) = postProject
+		{E5606ECE-48CA-4464-BB12-09D81D02B9EF} = {E5606ECE-48CA-4464-BB12-09D81D02B9EF}
+	EndProjectSection
+EndProject
+Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "selectivesearch", "selectivesearch", "{BEF04803-47B4-4322-B9D7-E10A8468E79F}"
+	ProjectSection(SolutionItems) = preProject
+		Examples\Image\Detection\FastRCNN\selectivesearch\__init__.py = Examples\Image\Detection\FastRCNN\selectivesearch\__init__.py
+		Examples\Image\Detection\FastRCNN\selectivesearch\README.md = Examples\Image\Detection\FastRCNN\selectivesearch\README.md
+		Examples\Image\Detection\FastRCNN\selectivesearch\selectivesearch.py = Examples\Image\Detection\FastRCNN\selectivesearch\selectivesearch.py
+	EndProjectSection
+EndProject
+Project("{2150E333-8FDC-42A3-9474-1A3956D46DE8}") = "fastRCNN", "fastRCNN", "{C28E4FD7-F9A9-4473-8E5D-D209AF36A1E7}"
+	ProjectSection(SolutionItems) = preProject
+		Examples\Image\Detection\FastRCNN\fastRCNN\__init__.py = Examples\Image\Detection\FastRCNN\fastRCNN\__init__.py
+		Examples\Image\Detection\FastRCNN\fastRCNN\imdb.py = Examples\Image\Detection\FastRCNN\fastRCNN\imdb.py
+		Examples\Image\Detection\FastRCNN\fastRCNN\nms.py = Examples\Image\Detection\FastRCNN\fastRCNN\nms.py
+		Examples\Image\Detection\FastRCNN\fastRCNN\pascal_voc.py = Examples\Image\Detection\FastRCNN\fastRCNN\pascal_voc.py
+		Examples\Image\Detection\FastRCNN\fastRCNN\test.py = Examples\Image\Detection\FastRCNN\fastRCNN\test.py
+		Examples\Image\Detection\FastRCNN\fastRCNN\timer.py = Examples\Image\Detection\FastRCNN\fastRCNN\timer.py
+		Examples\Image\Detection\FastRCNN\fastRCNN\train_svms.py = Examples\Image\Detection\FastRCNN\fastRCNN\train_svms.py
+		Examples\Image\Detection\FastRCNN\fastRCNN\voc_eval.py = Examples\Image\Detection\FastRCNN\fastRCNN\voc_eval.py
+	EndProjectSection
+EndProject
 Global
 	GlobalSection(SolutionConfigurationPlatforms) = preSolution
 		Debug_CpuOnly|x64 = Debug_CpuOnly|x64
@@ -2037,6 +2085,16 @@ Global
 		{5D1972FA-F0A4-4035-8E63-8BAEF0230097}.Release_NoOpt|x64.Build.0 = Release_NoOpt|x64
 		{5D1972FA-F0A4-4035-8E63-8BAEF0230097}.Release|x64.ActiveCfg = Release|x64
 		{5D1972FA-F0A4-4035-8E63-8BAEF0230097}.Release|x64.Build.0 = Release|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Debug_CpuOnly|x64.ActiveCfg = Debug_CpuOnly|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Debug_CpuOnly|x64.Build.0 = Debug_CpuOnly|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Debug|x64.ActiveCfg = Debug|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Debug|x64.Build.0 = Debug|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Release_CpuOnly|x64.ActiveCfg = Release_CpuOnly|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Release_CpuOnly|x64.Build.0 = Release_CpuOnly|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Release_NoOpt|x64.ActiveCfg = Release_NoOpt|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Release_NoOpt|x64.Build.0 = Release_NoOpt|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Release|x64.ActiveCfg = Release|x64
+		{20DEE94F-2802-40B1-B88B-22755A03AA48}.Release|x64.Build.0 = Release|x64
 	EndGlobalSection
 	GlobalSection(SolutionProperties) = preSolution
 		HideSolutionNode = FALSE
@@ -2251,5 +2309,9 @@ Global
 		{C1189678-4FFA-4258-971F-3262B44FCA99} = {6994C86D-A672-4254-824A-51F4DFEB807F}
 		{F37067BD-8BB1-4F93-AEF4-F37434613AE4} = {DD043083-71A4-409A-AA91-F9C548DCF7EC}
 		{5D1972FA-F0A4-4035-8E63-8BAEF0230097} = {F37067BD-8BB1-4F93-AEF4-F37434613AE4}
+		{65649688-3377-4FA9-8CD0-BDC3AC72E0AD} = {3BF56127-6F0F-41CF-BFCE-31165A0A5E73}
+		{20DEE94F-2802-40B1-B88B-22755A03AA48} = {65649688-3377-4FA9-8CD0-BDC3AC72E0AD}
+		{BEF04803-47B4-4322-B9D7-E10A8468E79F} = {4EAFF1B2-2D70-4486-B95E-684E39A50609}
+		{C28E4FD7-F9A9-4473-8E5D-D209AF36A1E7} = {4EAFF1B2-2D70-4486-B95E-684E39A50609}
 	EndGlobalSection
 EndGlobal

CONTRIBUTING.md

@@ -1,7 +1,7 @@
 You want to contribute to CNTK? We're really excited to work together!
 
-Please, follow the steps from the Wiki Article at
+Please, follow the steps from the documentation:
 
-https://github.com/Microsoft/CNTK/wiki/Contributing-to-CNTK
+https://docs.microsoft.com/en-us/cognitive-toolkit/contributing-to-cntk
 
 Your CNTK team.

Dependencies/CNTKCustomMKL/README.md

@@ -6,8 +6,8 @@ for usage by CNTK ("CNTK custom MKL" for short).
 By default, a CNTK binary with Intel® MKL support includes a prebuilt CNTK
 custom MKL.
 If you want to build CNTK with Intel® MKL support yourself, you can install a
-prebuilt CNTK custom MKL, available for download from the [CNTK web site](https://www.cntk.ai/mkl).
-See [CNTK's setup instructions](https://github.com/Microsoft/CNTK/wiki/Setup-CNTK-on-your-machine)
+prebuilt CNTK custom MKL, available for download from the [here](https://www.microsoft.com/en-us/cognitive-toolkit/download-math-kernel-library/).
+See [CNTK's setup instructions](https://docs.microsoft.com/en-us/cognitive-toolkit/Setup-CNTK-on-your-machine)
 for more details.
 
 If you want to add new Intel® MKL functions to be used by CNTK you will have to

Documentation/CNTK-TechReport/lyx/CNTKBook_ASRDecoder_Chapter.lyx

@@ -860,7 +860,7 @@ The decoder uses a Viterbi beam search algorithm, in which unlikely hypotheses
  are pruned at each frame.
  The -beam parameter prevents unlikely hypotheses from being pursued.
  Any hypothesis that differs from the best hypothesis by more than this
- amount will be be discarded.
+ amount will be discarded.
  The -max-tokens parameter controls the number of active hypotheses.
  If the -beam parameter causes more than max-tokens hypotheses to be generated,
  then only this many of the best hypotheses will be retained.

Documentation/CNTK-TechReport/lyx/CNTKBook_CN_Chapter.lyx

@@ -10194,7 +10194,7 @@ reference "fig:CN-WithDelayNode"
 .
  Different from the CN without a directed loop, a CN with a loop cannot
  be computed for a sequence of samples as a batch since the next sample's
- value depends on the the previous samples.
+ value depends on the previous samples.
  A simple way to do forward computation and backpropagation in a recurrent
  network is to unroll all samples in the sequence over time.
  Once unrolled, the graph is expanded into a DAG and the forward computation

Documentation/CNTK-TechReport/lyx/CNTKBook_ExampleSetup_Chapter.lyx

@@ -1505,7 +1505,7 @@ In this script, the initial model is loaded and set as the default model.
  Note that through the use of configuration variables, this same script
  can be reused anytime a new layer can be added.
  If different layer sizes were desired, the HDim variable could be set by
- the higher level configuration file in the the appropriate edit block,
+ the higher level configuration file in the appropriate edit block,
  rather than from within the MEL script.
 
 \end_layout