Difference between revisions of "Machine Learning: Linear Regression"

From APL
Jump to: navigation, search
(Created page with "Machine learning is the subfield of Artificial intelligence interested in making agents that review their past actions and experiences to formulate a new action. The task is t...")
 
Line 2: Line 2:
  
 
Here are the packages that are used and three points to approximate a function from.  
 
Here are the packages that are used and three points to approximate a function from.  
 +
 +
 
<code>
 
<code>
 
import numpy as np
 
import numpy as np
Line 12: Line 14:
  
 
The slope and intercept must be initialized by the user. In addition the learning rate is how quickly to change the parameters by each iteration. the data set is fitted multiple times with updated starting weights, these are epochs.  
 
The slope and intercept must be initialized by the user. In addition the learning rate is how quickly to change the parameters by each iteration. the data set is fitted multiple times with updated starting weights, these are epochs.  
 +
  
 
<code>
 
<code>
Line 36: Line 39:
 
         m=m-alpha*error*x
 
         m=m-alpha*error*x
 
         b=b-alpha*error
 
         b=b-alpha*error
 
 
     print(m,b)
 
     print(m,b)
 
</code>
 
</code>
  
the slope is found to be 1.04 and the y-intercept as -0.08. The exact values were not found because this algorithm approximates the local minimum. It overshoots the exact value and then updates it by overshooting it again resulting in the answer revolving around the exact value.
+
the slope is found to be 1.04 and the y-intercept as -0.08. Imperfect data was used so the exact value was not found. The choice of alpha is important because a value that is too large will give an erroneous result. The algorithm will overshoot the exact value and then update it by overshooting it again resulting in the answer revolving around the exact value.
 +
 
 +
This is a graph of the final parameters:
 +
[[File:Linearregression2tut.png]]

Revision as of 00:04, 12 November 2016

Machine learning is the subfield of Artificial intelligence interested in making agents that review their past actions and experiences to formulate a new action. The task is to learn the formula for a function. A performance standard is used to adjust the function that is predicted. Linear regression is a common pattern that can be approximated by learning the parameters for slope and intercept.

Here are the packages that are used and three points to approximate a function from.


import numpy as np import matplotlib.pyplot as plt pair=np.array([(1,1),(4,4.2),(2,2)])

Scatterlinearpost.png

The slope and intercept must be initialized by the user. In addition the learning rate is how quickly to change the parameters by each iteration. the data set is fitted multiple times with updated starting weights, these are epochs.


nepoch = 4 m=0.0 b=0.0 alpha = 0.2

each pair is looped through and a y is predicted according to Y_pred=X*m+b The error is then found according to Y_pred-Y The weights are then updated according to the formula m:=m-learning_rate*error*x and b:=b-learning_rate*error The new value of the paramaters are used with the next set of points.

for i in range(0,nepoch,1):

   print("i=" , i)
   for i in [0,1,2]:
       print(m,b)
       x=pair[i][0]
       y=pair[i][1]
       y_pred=m*x+b
       error=y_pred-y
       m=m-alpha*error*x
       b=b-alpha*error
   print(m,b)

the slope is found to be 1.04 and the y-intercept as -0.08. Imperfect data was used so the exact value was not found. The choice of alpha is important because a value that is too large will give an erroneous result. The algorithm will overshoot the exact value and then update it by overshooting it again resulting in the answer revolving around the exact value.

This is a graph of the final parameters: Linearregression2tut.png