Introduction To Machine Learning In Node.js

How to combine machine learning code in your Node.js server

Transcript

1. Introduction To Machine Learning In Node.js Tamar Tena-Stern
2. Tamar Twena-Stern • Software Engineer - manager and architect • Architect @PaloAltoNetworks • Was a CTO of my own startup • Passionate about Node.js ! • Twitter: @SternTwena
3. Tamar Twena-Stern • Just Finished My Maternity Leave • Have 3 kids • Loves to play my violin • Javascript Israel community leader
4. What We Can Do With Machine Learning ?
5. What is Machine Learning • Give training data to your model • Your model implement an algorithm • The model study • Patterns on your data • Inference on your data • Use those patterns to perform a task with no instructions
6. Some Machine Learning Applications
7. Machine Learning Fundamentals
8. How To Build Machine Learning Model • Define - What type of model will it be • Fit - Capture patterns from provided data • Predict - Just what it sounds like • Evaluate - Determine how accurate the model's predictions are.
9. An Example For A DataSet
10. Step 1 - Choosing A Model To Train • Model #1 - Decision Tree Model • Easier to understand • The basic for more complicated Machine Learning Algorithms • Lets assume we predict house prices
11. Lets Predict House Prices With Simple Decision Tree
12. More Sophisticated Decision Tree
13. Step 2 - Lets Fit And Predict The Data • Select several columns from the dataset as variable X • Select one column of the data set that we want to predict • Called Prediction Target - Marked as Y • Predict patterns and according to them give results on new data
14. Node.js Decision Tree Demo
15. Step 3 - Evaluate How Accurate Are The Predictions Of Your Model Error = abs(actual−predicted) Number Of Testing Subjects
16. Measuring Errors Demo
17. Popular Node.js Machine Learning Libraries • TensorFlow.js • Brain.js • Synaptic.js • ConvNetJS
18. TensorFlow.js • Open-source • Training and deploying machine learning models. • Main operations relates to vector and matrices • Basic unit - Tensor
19. Demo - Image Classification With TensorFlow
20. Brain.js
21. Brain.js In A Nutshell • GPU accelerated library of Neural Networks • Fast and easy to learn • You do not have to know neural networks in order to work with it
22. Neural Networks In A Nutshell • Set of Algorithms • Modelled loosely after the human brain • Designed to recognise patterns
23. What Is A Neuron
24. Basic Unit Of Neural Networks - Artificial Neurons
25. How The Network Learns ? Blue Blue Yellow
26. Brain.js Demo
27. Did you notice that the algorithm was CPU intensive ?
28. Is JavaScript The Go-To Programming Language For Machine Learning ?
30. Main Problems • Machine Learning algorithms are CPU intensive • Matrix manipulations can be more efficient • Not much richness of libraries
31. Lets Deep Dive Into The CPU Intensive Problem
32. Node.js Architecture
33. Why CPU Intensive Operations Don’t Shine In Node • Non Blocking IO + Event Loop • Constant amour of threads • CPU Intensive operation will block the event loop • CPU intensive operation that will be off loaded to the worker thread pool will make one of the threads busy for long
34. Demo - CPU Intensive
35. CPU Intensive Tasks In Node.js • Fork Another Process • Worker Threads • Scale Up with Queues + pack our application with containers
36. Child Process
37. Prefer Not To Use Child Process • Server running on specific hardware • Bounded amount of CPU • Usually , each process will be mapped to a CPU • Multiple requests to your server - many processes • Will not scale
38. Worker Threads • Available since 10.5.0 • Enable the use of threads to execute JavaScript in Parallel • One process, Multiple threads, one event loop per thread, One JS engine per thread, one Node.js instance per thread
39. Demo - Brain.js With Worker Threads
40. Scale Up With Queues
41. Queues VS Worker Threads • Queues Are Better for scale up • Worker threads limit you to the current machine’s resources • Can use Multiple CPUs from multiple machines while scaling up
42. Questions ?
43. • Twitter: @SternTwena