Support Vector Machines non probabilistic binary linear classifier
There is a decision boundary
When y=0 and Theta dot x <= -1, cost is 0
When y=1 and Theta dot x >= 1, cost is 0
Best line is decision boundary that maximizes distance to the positive and maximize distance to negative
Nomral of the decision boundary maximies lengths of p means it is hitting source of light the most effectively
There is a decision boundary
When y=0 and Theta dot x <= -1, cost is 0
When y=1 and Theta dot x >= 1, cost is 0
Best line is decision boundary that maximizes distance to the positive and maximize distance to negative
Nomral of the decision boundary maximies lengths of p means it is hitting source of light the most effectively
Comments
Post a Comment