Scribd
Upload
33 views10 pages

Linear Regression

Uploaded by

Sanskriti Gupta
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
33 views10 pages

Linear Regression

Uploaded by

Sanskriti Gupta
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

Linear Regression

Regression
It is the statistical technique that relates a dependant variable to one or more independant variables

training set

learning algo

x
f y^
input/feature (function)(model) predicted value

w and b : parameters
coefficients
weights
Cost Function

error -> y^(i) - y(i)


squared error cost function:

y(i)
error

y^(i)

goal is to minimize J(w,b) hence minimising the error


Cost Function Intuition

w = 0.5

w=
- 0.
5
simplified: b=0
=
f(x) = wx
Gradient Descent
0<alpha<1 number of steps downhill
gives direction

if alpha is too small : if alpha is too large:


gradient descent may be too slow gradient descent may never reach min & overshoot
Gradient Descent Implementation
Multi Variable Linear Regression
Gradient Descent

You might also like