[go: up one dir, main page]
Scribd
Upload
3 views13 pages

Section+04+ +classification+ +003+ +SVM

The document provides an overview of Support Vector Machines (SVM), a supervised learning algorithm primarily used for classification tasks. It explains the concept of hyperplanes that separate classes, the importance of selecting the right hyperplane, and the factors involved in training a two-class SVM using Azure ML. Key parameters for training include iterations, regularization, feature normalization, and handling unknown categories.

Uploaded by

sundar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views13 pages

Section+04+ +classification+ +003+ +SVM

The document provides an overview of Support Vector Machines (SVM), a supervised learning algorithm primarily used for classification tasks. It explains the concept of hyperplanes that separate classes, the importance of selecting the right hyperplane, and the factors involved in training a two-class SVM using Azure ML. Key parameters for training include iterations, regularization, feature normalization, and handling unknown categories.

Uploaded by

sundar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

© Jitesh Khurkhuriya – Azure ML Online Course

Classification

© Jitesh Khurkhuriya – Azure ML Online Course


Support Vector Machine

© Jitesh Khurkhuriya – Azure ML Online Course


What is SVM?
• Supervised Learning Algorithm

• Can be used for both Regression as well as Classification

• Mostly used for classification

• The observations are separated by a hyperplane in the space

© Jitesh Khurkhuriya – Azure ML Online Course


Vectors

X2

X1

© Jitesh Khurkhuriya – Azure ML Online Course


Vectors
X1

P1 (x1,x2,x3)

X2

X3

© Jitesh Khurkhuriya – Azure ML Online Course


Hyperplane

© Jitesh Khurkhuriya – Azure ML Online Course


Hyperplane

X2

X1

A hyperplane separates the two classes.


© Jitesh Khurkhuriya – Azure ML Online Course
Choosing a Hyperplane

X2

X1

Select a hyperplane that separates the two classes.


© Jitesh Khurkhuriya – Azure ML Online Course
Select the right hyperplane
A Hyperplanes
X2 C
B

Margin Support vectors

X1

Margin is maximum distance between the nearest data points and the hyperplane.
© Jitesh Khurkhuriya – Azure ML Online Course
What comes first?
A
X2
B

X1

Identifying the accurate classes comes first before margin calculation.


© Jitesh Khurkhuriya – Azure ML Online Course
Two-Class SVM using Azure ML
• Create Trainer Mode
• Single Parameter – accepts a specific set of values
• Parameter Range – Finds optimal parameter when used with Hyper Parameter Tuning

• Number of Iterations

• Lambda – Value to use for L1 Regularisation. Larger value penalises the model

• Normalize Features – Training data sets are centred at the mean and scaled to have one unit of
standard deviation

• Project to the unit sphere – to normalize the coefficients

• Random Number Seed – any integer value for reproducing the results

• Allow unknown category - Creates a group for unknown values in the training or validation sets.

© Jitesh Khurkhuriya – Azure ML Online Course


Thank You…!

© Jitesh Khurkhuriya – Azure ML Online Course

You might also like