think machine-learning-with-scikit-learn-chetan

49
Source code: https://github.com/dskskv/Think-ML/

Upload: chetan-khatri

Post on 21-Jan-2018

347 views

Category:

Data & Analytics


4 download

TRANSCRIPT

Source code: https://github.com/dskskv/Think-ML/

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

Early Days AI Programs : Deep Blue

Now, AI Programs

● Alpha go is best example, wrote for Playing Go game, but it can play Atari games also.

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”

Let's have problem

Let's have problem: It seems easy but difficult to solve without machine learning.

Open Source Libraries

Classifier

Scikit-learn

Test ! No error ! Yay !!

Supervised Learning

Collecting Training

Data

Train Classifier

MakePredictions

Training DataWeight Texture Label150g Bumpy Orange170g Bumpy Orange140g Smooth Apple130g Smooth Apple

Features

Examples

Training Data

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 ?

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..

Demo

2. Visualizing a Decision Tree

3. What Makes a Good Feature?

Imagine we want to write classifier to classify two types of dogs.

Variation in the world !

Hands - On Session

https://github.com/dskskv/Think-ML/

About 80% of dogs at this height are labs

About 95% of dogs at this height are greyhounds

lFeature captures different types of information

Thought Experiment

Avoid useless features

Independent features are best

Height in InchesHeight in centimeters

Avoid Redundant featuresFeature should be easy to understand

Thank you

@khatri_chetan