Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Upcoming SlideShare
Exploring Machine Learning in Python with Scikit-Learn
Next
Download to read offline and view in fullscreen.

2

Share

Download to read offline

Think machine-learning-with-scikit-learn-chetan

Download to read offline

A Handson session on Machine learning with Scikit-learn

Think machine-learning-with-scikit-learn-chetan

  1. 1. Source code: https://github.com/dskskv/Think-ML/
  2. 2. Outline ● An Introduction to Machine Learning ● Hello World in Machine learning with 6 lines of code ● Visualizing a Decision Tree ● Classifying Images ● Supervised learning : Pipeline ● Writing first Classifier
  3. 3. Early Days AI Programs : Deep Blue
  4. 4. Now, AI Programs ● Alpha go is best example, wrote for Playing Go game, but it can play Atari games also.
  5. 5. Machine Learning Machine Learning does this possible, it is study of algorithms which learns from examples and experience having set of rules and hardcoded lines. “Learns from Examples and Experience”
  6. 6. Let's have problem Let's have problem: It seems easy but difficult to solve without machine learning.
  7. 7. Open Source Libraries
  8. 8. Classifier
  9. 9. Scikit-learn
  10. 10. Test ! No error ! Yay !!
  11. 11. Supervised Learning Collecting Training Data Train Classifier Make Predictions
  12. 12. Training Data Weight Texture Label 150g Bumpy Orange 170g Bumpy Orange 140g Smooth Apple 130g Smooth Apple Features Examples
  13. 13. Training Data
  14. 14. Important Concepts ● How does this work in Real world ? ● How much training data do you need ? ● How is the tree created ? ● What makes a good feature ?
  15. 15. Many Types of Classifier ● Artificial Neural Network (ANN) ● Support Vector Machine (SVM) ● Nearest Neighbour classifier (KNN) ● Random Forest (RF) ● Gradient Boosting Machine (GBM) ● Etc.. ● Etc..
  16. 16. Demo
  17. 17. 2. Visualizing a Decision Tree
  18. 18. 3. What Makes a Good Feature? Imagine we want to write classifier to classify two types of dogs.
  19. 19. Variation in the world !
  20. 20. Hands - On Session https://github.com/dskskv/Think-ML/
  21. 21. About 80% of dogs at this height are labs
  22. 22. About 95% of dogs at this height are greyhounds
  23. 23. lFeature captures different types of information
  24. 24. Thought Experiment
  25. 25. Avoid useless features
  26. 26. Independent features are best
  27. 27. Height in Inches Height in centimeters
  28. 28. Avoid Redundant features Feature should be easy to understand
  29. 29. Thank you @khatri_chetan
  • DanishNazer

    Jun. 20, 2017
  • ThomasHormaza

    Jun. 13, 2017

A Handson session on Machine learning with Scikit-learn

Views

Total views

730

On Slideshare

0

From embeds

0

Number of embeds

1

Actions

Downloads

31

Shares

0

Comments

0

Likes

2

×