This is the slidedeck used for the presentation of the Amsterdam Pipeline of Data Science, held in December 2016. TensorFlow in the open source library from Google to implement deep learning, neural networks. This is an introduction to Tensorflow. Note: Videos are not included (which were shown during the presentation)

1. Introducing Tensorflow
The game changer in building "intelligent" applications

2. Who Am I
• Rokesh Jankie (Computer Science, MSc)
• Google Believer since Gmail (2004)
• Professionally :
• CTO QAFE Inc., Head of R&D Qualogy
• Other:
• Organizer for GDG Netherlands and GDG
Cloud Netherlands
• Was introduced to Neural Networks in 1997

3. Shift in Strategy at Google

4. 4th of October 2016

5. 4th of October 2016

6. How Can You Get Started with Machine Learning?
• Three ways, with varying complexity:
• Use a Cloud-based or Mobile API (Vision,
Natural Language, etc.)
• Use an existing model architecture, and retrain
it or fine tune on your dataset
• Develop your own machine learning models for
new problems
Moreflexible,butmoreeffortrequired

8. What is TensorFlow
• Open source Machine Learning library
• Especially useful for Deep Learning
• For research and production
• Apache 2.0 license

9. TensorFlow History
• DistBelief

10. Jeff Dean!

11. You mentioned Deep Learning? Wazup?
• The first question to answer: What’s a Neural Network ?
• Inspired by Biology:
• Two flavours: Supervised an Unsupervised

12. Base Idea of Neural Networks

13. This is what we are trying to solve
f(x) = y

14. Activation Functions

15. Neural Networks
• Many kinds op NNs

16. What happened in the last decade…
• Algorithms: This area has seen some improvements, but most of the early
wins came from fairly old ideas. Now that Deep Learning is showing success
we are seeing some good advances as well.
• Datasets: Training large networks is hard without large enough datasets.
MNIST can only go so far in pushing the limits. Having datasets like ImageNet
has really helped pushed the state of the art in vision.
• Compute: the biggest game changer in recent years.

17. Convolution NNs
• Large Networks
• Lots of Input Nodes
• Lots of Layers
• Each Layer with a special task

18. TensorFlow History: The Cat
This was in 2012
It took 16,000
computers to
identify a cat!

19. “TensorFlow”, the name…
Tensors: multidimensional data arrays
[ 0.3, 0.2, 0.4, 0.5, 0.9 ]
But also:
[
[ 0.3, 0.2, 0.4, 0.5, 0.9 ],
[ 0.6, 0.0, 0.8, 0.5, 0.4 ],
[ 0.2, 0.8, 0.8, 0.1, 0.3 ]
]

20. “TensorFlow”, the flow ?
• Math Operations
• tf.exp tf.pow tf.tan tf.sign
• Control Flow
• tf.cond tf.do_while
• Tensor Operations
• tf.matmul tf.add tf.reduce_sum tf.cumprod

21. “TensorFlow”, the Graph
Computation is defined as a directed acyclic graph (DAG) to optimize an
objective function
• Graph is defined in high-level language (Python)
• Graph is compiled and optimized
• Graph is executed (in parts or fully) on available low level devices (CPU, GPU)
• Data (tensors) flow through the graph
• TensorFlow can compute gradients automatically

22. “TensorFlow”, the Graph

23. TensorBoard Projector (new)
Recently launched
Projector

24. Note on t-SNE
The good news is that by studying how t-SNE behaves in simple
cases, it’s possible to develop an intuition for what’s going on.
(see also http://distill.pub/2016/misread-tsne/ )

25. “TensorFlow”, Gradients?

26. The architecture
Core in C++
Different front ends
Python and C++ today, community may add more

28. Linear Regression & Classification

29. TensorFlow Playground

31. what just happened in Linear Regression ?

32. Writing your Applications:
Where to start ?
• The easiest way to go (nowadays) is Docker image
• When you start the docker image immediately a iPython Notebook is
available
• Start playing around with notebooks! (we can do it together!)

33. Code Sample 1: Simple

34. Code Sample 2: Which processing unit?

35. Code Sample 2 : output

36. Code Sample 2 : result

37. Code Sample 3: Simple NN

38. What is the process ?

39. Cloud ML
Announced in October 2016 at Google Horizon Event

40. The Power of Cloud ML

41. Examples
• https://github.com/tensorflow/
models/tree/master/im2txt
• “Lip reading”
• Human-like sound: WaveNet

42. Possibilities…many, but I want to
highlight one specifically

44. Solution space

45. – Aja Huang
“The possibilities and power are innumerable”

46. What’s next ?
• tensorflow.org
• github.com/tensorflow
• Want to learn more?
• Udacity class on Deep Learning, goo.gl/iHssII
• Guides, codelabs, videos
• MNIST for Beginners, goo.gl/tx8R2b
• TF Learn Quickstart, goo.gl/uiefRn
• TensorFlow for Poets, goo.gl/bVjFIL
• ML Recipes, goo.gl/KewA03
• TensorFlow and Deep Learning without a PhD,
goo.gl/pHeXe7

47. And there is more
• GDG Cloud Netherlands is organizing a Viewing Party @Google Amsterdam
• More TensorFlow coming up (hands-on, more in-depth session, training)
Follow twitter: @gdgcloudnl
subscribe to https://www.gdgnetherlands.org

48. Let’s create awesome stuff!Rokesh Jankie
rjankie@gmail.com
twitter.com/rjankie
plus.google.com/+RokeshJankie/
Many thanks to Josh Gordon, Yann LeCunn, Chris Olah, Jeff Dean