Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
SlideShare a Scribd company logo

Microsoft Cognitive Toolkit (Atlanta Code Camp 2017)

•
•
Mark Tabladillo
Mark Tabladillo

The Microsoft Cognitive Toolkit (CNTK) is a unified deep-learning toolkit that describes neural networks as a series of computational steps via a directed graph. In this directed graph, leaf nodes represent input values or network parameters, while other nodes represent matrix operations upon their inputs. The objectives of this presentation is to 1) describe what CNTK is, 2) present a comparative evaluation with similar technologies, 3) outline potential applications, and 4) demonstrate the technology with Jupyter Python examples.

Related slideshows

Testing for the deeplearning folks
Testing for the deeplearning folksTesting for the deeplearning folks
Testing for the deeplearning folks
 
Approximate "Now" is Better Than Accurate "Later"
Approximate "Now" is Better Than Accurate "Later"Approximate "Now" is Better Than Accurate "Later"
Approximate "Now" is Better Than Accurate "Later"
 
H2O & Tensorflow - Fabrizio
H2O & Tensorflow - Fabrizio H2O & Tensorflow - Fabrizio
H2O & Tensorflow - Fabrizio
 
1 of 30
Download to read offline
Microsoft Cognitive Toolkit (Atlanta Code Camp 2017)
With many contributors: A. Agarwal, E. Akchurin, E. Barsoum, C. Basoglu, G. Chen, S. Cyphers, W. Darling, J. Droppo, A. Eversole, B. Guenter, M. Hillebrand, X. Huang, Z. Huang, R. Hoens, V. Ivanov, A. Kamenev, N. Karampatziakis, P. Kranen, O. Kuchaiev, W. Manousek, C. Marschner, A. May, B. Mitra, O. Nano, G. Navarro, A. Orlov, S. Pathak, M. Radmilac, A. Reznichenko, P. Parthasarathi, B. Peng, A. Reznichenko, W. Richert, F. Seide, M. Seltzer, M. Slaney, A. Stolcke, T. Will, H. Wang, W. Xiong, K. Yao, D. Yu, C. Zhang, Y. Zhang, G. Zweig
Objectives • Describe the Microsoft Cognitive Toolkit (CNTK) • Comparative evaluation with similar technologies • Outline potential applications • Show Jupyter Python demos
Assumptions • Intermediate knowledge of programming • Intermediate knowledge of data science or machine learning • Stay Calm: You may fill in gaps later
What is Cognitive Toolkit (CNTK)
The Microsoft Cognitive Toolkit (CNTK) • Microsoft’s open-source deep-learning toolkit • ease of use: what, not how • fast • flexible • Supported on both Linux and Windows • Fully open development framework
• CNTK is production-ready: State-of-the-art accuracy, efficient, and scales to multi-GPU/multi-server. • CNTK expresses (nearly) arbitrary neural networks by composing simple building blocks into complex computational networks, supporting relevant network types and applications. The Microsoft Cognitive Toolkit (CNTK)
Microsoft’s historic speech breakthrough • Microsoft 2016 research system for conversational speech recognition • 5.9% word-error rate • enabled by CNTK’s multi-server scalability [W. Xiong, J. Droppo, X. Huang, F. Seide, M. Seltzer, A. Stolcke, D. Yu, G. Zweig: “Achieving Human Parity in Conversational Speech Recognition,” https://arxiv.org/abs/1610.05256]
Microsoft’s historic speech breakthrough • Microsoft 2017 research system for conversational speech recognition • 5.1% word-error rate • “We introduced an additional CNN-BLSTM (convolutional neural network combined with bidirectional long-short-term memory) model for improved acoustic modeling." • Additionally, Microsoft enlisted its suite of AI technologies called the Cognitive Toolkit (CNTK) in its record-setting bid.
Microsoft Cognitive Toolkit (Atlanta Code Camp 2017)
Comparative Evaluation
28.2 25.8 16.4 11.7 7.3 6.7 3.6 0.0 5.0 10.0 15.0 20.0 25.0 30.0 ILSVRC 2010 NEC America ILSVRC 2011 Xerox ILSVRC 2012 AlexNet ILSVRC 2013 Clarifi ILSVRC 2014 VGG ILSVRC 2014 GoogleNet ILSVRC 2015 ResNet ImageNet Classification top-5 error (%) Microsoft had all 5 entries being the 1-st places in 2015: ImageNet classification, ImageNet localization, ImageNet detection, COCO detection, and COCO segmentation
CNTK Benchmarked by HKBU (September 2016) • CNTK’s LSTM performance is 5-10x faster than the other toolkits. • For convolution (image tasks), CNTK is comparable, but note the authors were using CNTK 1.7.2, and current CNTK 2.0 beta 10 is over 30% faster than 1.7.2. • For all networks, CTNK's performance was superior to TensorFlow performance.
Benchmarking on a single server by HKBU – September 2016 “CNTK is production-ready: State-of-the-art accuracy, efficient, and scales to multi-GPU/multi-server.” FCN-8 AlexNet ResNet-50 LSTM-64 CNTK 0.037 0.040 (0.054) 0.207 (0.245) 0.122 Caffe 0.038 0.026 (0.033) 0.307 (-) - TensorFlow 0.063 - (0.058) - (0.346) 0.144 Torch 0.048 0.033 (0.038) 0.188 (0.215) 0.194 G980
Frames Processing Source: https://github.com/Microsoft/CNTK/blob/master/Documentation/Documents/PerformanceChart.png • The figure compares processing speed (frames processed per second) of CNTK to that of four other well-known toolkits. • The configuration uses a fully connected 4-layer neural network (see our benchmark scripts) and an effective mini batch size (8192). • All results were obtained on the same hardware with the respective latest public software versions as of Dec 3, 2015.
Benchmarking CNTK on Keras: is it Better at Deep Learning than TensorFlow? • Deciding the better Keras framework is not as clear cut as expected. • Accuracy is mostly identical between the two frameworks. CNTK is faster at LSTMs/MLPs, TensorFlow is faster at CNNs/Embeddings • When networks implement both, it’s a tie. http://minimaxir.com/2017/06/keras-cntk/
Potential Applications
Deep learning at Microsoft • Microsoft Cognitive Services • Skype Translator • Cortana • Bing • Bing Ads • Augmented Reality • Microsoft Research
Intelligent Kiosk Demo
Intelligent Kiosk • https://github.com/Microsoft/Cognitive-Samples- IntelligentKiosk
CNTK Code
Jupyter http://jupyter.org/
Github • Equifax may have given away my birth date, my social security number, and my credit cards • But they’re not getting into my Github
• Github: https://github.com/Microsoft/CNTK • Wiki: https://github.com/Microsoft/CNTK/wiki • Issues: https://github.com/Microsoft/CNTK/issues Microsoft Cognitive Toolkit
Microsoft Cognitive Toolkit Examples https://github.com/Microsoft/CNTK
Microsoft MML Spark
Jupyter Python Demos
Apache Spark • SPARK2017 https://www.wintellectnow.com/Home/Instructor?instructorId=Mark_Tabladillo
Summary of CNTK • Microsoft’s open-source deep-learning toolkit • Supported on both Linux and Windows • Fully open development framework
Abstract The Microsoft Cognitive Toolkit (CNTK) is a unified deep-learning toolkit that describes neural networks as a series of computational steps via a directed graph. In this directed graph, leaf nodes represent input values or network parameters, while other nodes represent matrix operations upon their inputs. The objectives of this presentation is to 1) describe what CNTK is, 2) present a comparative evaluation with similar technologies, 3) outline potential applications, and 4) demonstrate the technology with Jupyter Python examples.

More Related Content

Microsoft Cognitive Toolkit (Atlanta Code Camp 2017)

  • 2. With many contributors: A. Agarwal, E. Akchurin, E. Barsoum, C. Basoglu, G. Chen, S. Cyphers, W. Darling, J. Droppo, A. Eversole, B. Guenter, M. Hillebrand, X. Huang, Z. Huang, R. Hoens, V. Ivanov, A. Kamenev, N. Karampatziakis, P. Kranen, O. Kuchaiev, W. Manousek, C. Marschner, A. May, B. Mitra, O. Nano, G. Navarro, A. Orlov, S. Pathak, M. Radmilac, A. Reznichenko, P. Parthasarathi, B. Peng, A. Reznichenko, W. Richert, F. Seide, M. Seltzer, M. Slaney, A. Stolcke, T. Will, H. Wang, W. Xiong, K. Yao, D. Yu, C. Zhang, Y. Zhang, G. Zweig
  • 3. Objectives • Describe the Microsoft Cognitive Toolkit (CNTK) • Comparative evaluation with similar technologies • Outline potential applications • Show Jupyter Python demos
  • 4. Assumptions • Intermediate knowledge of programming • Intermediate knowledge of data science or machine learning • Stay Calm: You may fill in gaps later
  • 5. What is Cognitive Toolkit (CNTK)
  • 6. The Microsoft Cognitive Toolkit (CNTK) • Microsoft’s open-source deep-learning toolkit • ease of use: what, not how • fast • flexible • Supported on both Linux and Windows • Fully open development framework
  • 7. • CNTK is production-ready: State-of-the-art accuracy, efficient, and scales to multi-GPU/multi-server. • CNTK expresses (nearly) arbitrary neural networks by composing simple building blocks into complex computational networks, supporting relevant network types and applications. The Microsoft Cognitive Toolkit (CNTK)
  • 8. Microsoft’s historic speech breakthrough • Microsoft 2016 research system for conversational speech recognition • 5.9% word-error rate • enabled by CNTK’s multi-server scalability [W. Xiong, J. Droppo, X. Huang, F. Seide, M. Seltzer, A. Stolcke, D. Yu, G. Zweig: “Achieving Human Parity in Conversational Speech Recognition,” https://arxiv.org/abs/1610.05256]
  • 9. Microsoft’s historic speech breakthrough • Microsoft 2017 research system for conversational speech recognition • 5.1% word-error rate • “We introduced an additional CNN-BLSTM (convolutional neural network combined with bidirectional long-short-term memory) model for improved acoustic modeling." • Additionally, Microsoft enlisted its suite of AI technologies called the Cognitive Toolkit (CNTK) in its record-setting bid.
  • 12. 28.2 25.8 16.4 11.7 7.3 6.7 3.6 0.0 5.0 10.0 15.0 20.0 25.0 30.0 ILSVRC 2010 NEC America ILSVRC 2011 Xerox ILSVRC 2012 AlexNet ILSVRC 2013 Clarifi ILSVRC 2014 VGG ILSVRC 2014 GoogleNet ILSVRC 2015 ResNet ImageNet Classification top-5 error (%) Microsoft had all 5 entries being the 1-st places in 2015: ImageNet classification, ImageNet localization, ImageNet detection, COCO detection, and COCO segmentation
  • 13. CNTK Benchmarked by HKBU (September 2016) • CNTK’s LSTM performance is 5-10x faster than the other toolkits. • For convolution (image tasks), CNTK is comparable, but note the authors were using CNTK 1.7.2, and current CNTK 2.0 beta 10 is over 30% faster than 1.7.2. • For all networks, CTNK's performance was superior to TensorFlow performance.
  • 14. Benchmarking on a single server by HKBU – September 2016 “CNTK is production-ready: State-of-the-art accuracy, efficient, and scales to multi-GPU/multi-server.” FCN-8 AlexNet ResNet-50 LSTM-64 CNTK 0.037 0.040 (0.054) 0.207 (0.245) 0.122 Caffe 0.038 0.026 (0.033) 0.307 (-) - TensorFlow 0.063 - (0.058) - (0.346) 0.144 Torch 0.048 0.033 (0.038) 0.188 (0.215) 0.194 G980
  • 15. Frames Processing Source: https://github.com/Microsoft/CNTK/blob/master/Documentation/Documents/PerformanceChart.png • The figure compares processing speed (frames processed per second) of CNTK to that of four other well-known toolkits. • The configuration uses a fully connected 4-layer neural network (see our benchmark scripts) and an effective mini batch size (8192). • All results were obtained on the same hardware with the respective latest public software versions as of Dec 3, 2015.
  • 16. Benchmarking CNTK on Keras: is it Better at Deep Learning than TensorFlow? • Deciding the better Keras framework is not as clear cut as expected. • Accuracy is mostly identical between the two frameworks. CNTK is faster at LSTMs/MLPs, TensorFlow is faster at CNNs/Embeddings • When networks implement both, it’s a tie. http://minimaxir.com/2017/06/keras-cntk/
  • 18. Deep learning at Microsoft • Microsoft Cognitive Services • Skype Translator • Cortana • Bing • Bing Ads • Augmented Reality • Microsoft Research
  • 23. Github • Equifax may have given away my birth date, my social security number, and my credit cards • But they’re not getting into my Github
  • 24. • Github: https://github.com/Microsoft/CNTK • Wiki: https://github.com/Microsoft/CNTK/wiki • Issues: https://github.com/Microsoft/CNTK/issues Microsoft Cognitive Toolkit
  • 29. Summary of CNTK • Microsoft’s open-source deep-learning toolkit • Supported on both Linux and Windows • Fully open development framework
  • 30. Abstract The Microsoft Cognitive Toolkit (CNTK) is a unified deep-learning toolkit that describes neural networks as a series of computational steps via a directed graph. In this directed graph, leaf nodes represent input values or network parameters, while other nodes represent matrix operations upon their inputs. The objectives of this presentation is to 1) describe what CNTK is, 2) present a comparative evaluation with similar technologies, 3) outline potential applications, and 4) demonstrate the technology with Jupyter Python examples.