LFD Book Forum

LFD Book Forum (http://book.caltech.edu/bookforum/index.php)
-   The Final (http://book.caltech.edu/bookforum/forumdisplay.php?f=138)
-   -   Q7 No feature transform but do we add an x_0 (http://book.caltech.edu/bookforum/showthread.php?t=1517)

patrickjtierney 09-14-2012 05:21 AM

Q7 No feature transform but do we add an x_0
 
Title says it all. Do we still add the column of 1's to X so that that the weight vector is of length 3, or not?

patrickjtierney 09-14-2012 06:20 AM

Re: Q7 No feature transform but do we add an x_0
 
I tried it both ways, and strangely enough, it turns out that the number of wrongly classified inputs in each case is exactly equal whether you use the X0=1 term or not, even though the weights change quite a lot. I wasn't expecting that.

Andrs 09-14-2012 07:20 AM

Re: Q7 No feature transform but do we add an x_0
 
Q7 is about plain linear regression and you should have a w0 parameter that corresponds to the constant bias.

yaser 09-14-2012 09:55 AM

Re: Q7 No feature transform but do we add an x_0
 
Quote:

Originally Posted by Andrs (Post 5264)
Q7 is about plain linear regression and you should have a w0 parameter that corresponds to the constant bias.

Correct.


All times are GMT -7. The time now is 06:12 PM.

Powered by vBulletin® Version 3.8.3
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
The contents of this forum are to be used ONLY by readers of the Learning From Data book by Yaser S. Abu-Mostafa, Malik Magdon-Ismail, and Hsuan-Tien Lin, and participants in the Learning From Data MOOC by Yaser S. Abu-Mostafa. No part of these contents is to be communicated or made accessible to ANY other person or entity.