1 / 80

Deep Learning with Google TensorFlow - A Comprehensive Course on Social Computing and Big Data Analytics

Join Tamkang University's Assistant Professor, Min-Yuh Day, in an in-depth course on Deep Learning with Google TensorFlow. Explore topics such as data science, big data analytics, map-reduce paradigm, deep learning, and social computing.

patriciaehernandez
Télécharger la présentation

Deep Learning with Google TensorFlow - A Comprehensive Course on Social Computing and Big Data Analytics

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. Tamkang University Tamkang University Deep Learning with Google TensorFlow (Google TensorFlow 深度學習) Social Computing and Big Data Analytics社群運算與大數據分析 1042SCBDA10 MIS MBA (M2226) (8628) Wed, 8,9, (15:10-17:00) (B309) Min-Yuh Day 戴敏育 Assistant Professor 專任助理教授 Dept. of Information Management, Tamkang University 淡江大學資訊管理學系 http://mail. tku.edu.tw/myday/ 2016-05-04

  2. 課程大綱 (Syllabus) 週次 (Week) 日期 (Date) 內容 (Subject/Topics) 1 2016/02/17 Course Orientation for Social Computing and Big Data Analytics (社群運算與大數據分析課程介紹) 2 2016/02/24 Data Science and Big Data Analytics: Discovering, Analyzing, Visualizing and Presenting Data (資料科學與大數據分析：探索、分析、視覺化與呈現資料) 3 2016/03/02 Fundamental Big Data: MapReduce Paradigm, Hadoop and Spark Ecosystem (大數據基礎：MapReduce典範、 Hadoop與Spark生態系統)

  3. 課程大綱 (Syllabus) 週次 (Week) 日期 (Date) 內容 (Subject/Topics) 4 2016/03/09 Big Data Processing Platforms with SMACK: Spark, Mesos, Akka, Cassandra and Kafka (大數據處理平台SMACK： Spark, Mesos, Akka, Cassandra, Kafka) 5 2016/03/16 Big Data Analytics with Numpy in Python (Python Numpy 大數據分析) 6 2016/03/23 Finance Big Data Analytics with Pandas in Python (Python Pandas 財務大數據分析) 7 2016/03/30 Text Mining Techniques and Natural Language Processing (文字探勘分析技術與自然語言處理) 8 2016/04/06 Off-campus study (教學行政觀摩日)

  4. 課程大綱 (Syllabus) 週次 (Week) 日期 (Date) 內容 (Subject/Topics) 9 2016/04/13 Social Media Marketing Analytics (社群媒體行銷分析) 10 2016/04/20 期中報告 (Midterm Project Report) 11 2016/04/27 Deep Learning with Theano and Keras in Python (Python Theano 和 Keras 深度學習) 12 2016/05/04 Deep Learning with Google TensorFlow (Google TensorFlow 深度學習) 13 2016/05/11 Sentiment Analysis on Social Media with Deep Learning (深度學習社群媒體情感分析)

  5. 課程大綱 (Syllabus) 週次 (Week) 日期 (Date) 內容 (Subject/Topics) 14 2016/05/18 Social Network Analysis (社會網絡分析) 15 2016/05/25 Measurements of Social Network (社會網絡量測) 16 2016/06/01 Tools of Social Network Analysis (社會網絡分析工具) 17 2016/06/08 Final Project Presentation I (期末報告 I) 18 2016/06/15 Final Project Presentation II (期末報告 II)

  6. Source: https://github.com/tensorflow/tensorflow

  7. Google TensorFlow https://www.tensorflow.org/

  8. TensorFlowis an Open Source Software Library for Machine Intelligence

  9. numerical computation using data flow graphs

  10. Nodes: mathematical operationsedges: multidimensional data arrays (tensors) communicated between nodes

  11. Computation is a Dataflow Graph Graph of Nodes, also called Operations or ops. bias weights Add Relu MatMul Xent examples labels Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  12. Computation is a Dataflow Graph Edges are N-dimensional arrays: Tensors bias weights Add Relu MatMul Xent inputs targets Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  13. Logistic Regression as Dataflow Graph Nodes Operations ops b bias W weights Add Softmax MatMul Xent X inputs Y targets Edges are N-dimensional arrays: Tensors Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  14. Computation is a Dataflow Graph with state ‘Biases’ is a variable Some ops compute gradients biases -= updates biases … Mul -= … Add learning rate Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  15. Neural Networks Input Layer (X) Hidden Layer (H) Output Layer (Y) X1 Y X2 Source: https://www.youtube.com/watch?v=bxe2T-V8XRs&index=1&list=PLiaHhY2iBX9hdHaRr6b7XevZtgZRa1PoU

  16. Data Flow Graph Source: https://www.tensorflow.org/

  17. Data Flow Graph Data Flow Graph Source: https://www.tensorflow.org/

  18. Data Flow Graph Data Flow Graph Source: https://www.tensorflow.org/

  19. TensorFlow Playground http://playground.tensorflow.org/

  20. TensorBoard https://www.tensorflow.org/tensorboard/index.html#graphs

  21. X Y Hours Sleep Hours Study Score 3 5 75 5 1 82 10 2 93 8 3 ? Source: https://www.youtube.com/watch?v=bxe2T-V8XRs&index=1&list=PLiaHhY2iBX9hdHaRr6b7XevZtgZRa1PoU

  22. X Y Hours Sleep Hours Study Score 3 5 75 Training 5 1 82 10 2 93 8 3 ? Testing Source: https://www.youtube.com/watch?v=bxe2T-V8XRs&index=1&list=PLiaHhY2iBX9hdHaRr6b7XevZtgZRa1PoU

  23. Training a Network=Minimize the Cost Function Source: https://www.youtube.com/watch?v=bxe2T-V8XRs&index=1&list=PLiaHhY2iBX9hdHaRr6b7XevZtgZRa1PoU

  24. Neural Networks Source: https://www.youtube.com/watch?v=bxe2T-V8XRs&index=1&list=PLiaHhY2iBX9hdHaRr6b7XevZtgZRa1PoU

  25. Neural Networks Input Layer (X) Hidden Layer (H) Output Layer (Y) X1 Y X2 Source: https://www.youtube.com/watch?v=bxe2T-V8XRs&index=1&list=PLiaHhY2iBX9hdHaRr6b7XevZtgZRa1PoU

  26. Neural Networks Input Layer (X) Hidden Layers (H) Output Layer (Y) Deep Neural Networks Deep Learning Source: https://www.youtube.com/watch?v=bxe2T-V8XRs&index=1&list=PLiaHhY2iBX9hdHaRr6b7XevZtgZRa1PoU

  27. Neural Networks Input Layer (X) Hidden Layer (H) Output Layer (Y) Neuron Synapse Synapse Neuron X1 Y X2 Source: https://www.youtube.com/watch?v=bxe2T-V8XRs&index=1&list=PLiaHhY2iBX9hdHaRr6b7XevZtgZRa1PoU

  28. Neuron and Synapse Source: https://en.wikipedia.org/wiki/Neuron

  29. Neurons 2 Bipolar neuron 1 Unipolar neuron 3 Multipolar neuron 4 Pseudounipolar neuron Source: https://en.wikipedia.org/wiki/Neuron

  30. Neural Networks Input Layer (X) Hidden Layer (H) Output Layer (Y) HoursSleep Score HoursStudy Source: https://www.youtube.com/watch?v=bxe2T-V8XRs&index=1&list=PLiaHhY2iBX9hdHaRr6b7XevZtgZRa1PoU

  31. Neural Networks Input Layer (X) Hidden Layer (H) Output Layer (Y) Source: https://www.youtube.com/watch?v=P2HPcj8lRJE&list=PLjJh1vlSEYgvGod9wWiydumYl8hOXixNu&index=2

  32. Convolutional Neural Networks (CNNs / ConvNets) http://cs231n.github.io/convolutional-networks/

  33. A regular 3-layer Neural Network http://cs231n.github.io/convolutional-networks/

  34. A ConvNet arranges its neurons in three dimensions (width, height, depth) http://cs231n.github.io/convolutional-networks/

  35. DeepDream Source: https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/tutorials/deepdream/deepdream.ipynb

  36. Try your first TensorFlow https://github.com/tensorflow/tensorflow

  37. Architecture of TensorFlow … C ++ front end Python front end Core TensorFlow Execution System CPU GPU Android iOS Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  38. Deep Learning A powerful class of machine learning model Modern reincarnation of artificial neural networks Collection of simple, trainable mathematical functions Compatible with many variants of machine learning Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  39. What is Deep Learning? Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  40. The Neuron x1 w1 w2 x2 y … … wn xn Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  41. The Neuron x1 w1 w2 x2 y … … wn xn Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  42. y = max ( 0, -0.21 * x1 + 0.3 * x2 + 0.7 * x3 ) Weights x1 -0.21 0.3 x2 y Inputs 0.7 x3 Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  43. Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  44. Learning Algorithm While not done: Pick a random training example “(input, label)” Run neural network on “input” Adjust weights on edges to make output closer to “label” Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  45. y = max ( 0, -0.21 * x1 + 0.3 * x2 + 0.7 * x3 ) Weights x1 -0.21 0.3 x2 y Inputs 0.7 x3 Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  46. Next time: y = max ( 0, -0.23 * x1 + 0.31 * x2 + 0.65 * x3 ) y = max ( 0, -0.21 * x1 + 0.3 * x2 + 0.7 * x3 ) Weights -0.23 x1 -0.21 0.31 0.3 x2 y Inputs 0.65 x3 0.7 Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  47. Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  48. Important Property of Neural Networks Results get better with More data + Bigger models + More computation (Better algorithms, new insights and improved techniques always help, too!) Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  49. Source: Jeff Dean (2016), Large-Scale Deep Learning For Building Intelligent Computer Systems, WSDM 2016

  50. Install TensorFlow https://www.tensorflow.org/versions/r0.8/get_started/os_setup.html

More Related