0

I am taking a machine learning class and I am struggling mightily to implement the algorithms (I'm an EE major). I have been given a set of 400 vectors, 200 for each classifier. I need to train the network for classify the 2-D vectors.

It would be a great help if I could be pointed in the right direction!

Logg
  • 1
  • This topic is to broad. There is only one direction with all possible explanations(scroll to the bottom to check the algorithm) http://www.mathworks.com/help/stats/kmeans.html – madbitloman Mar 25 '15 at 23:18
  • I believe this basically gives the centroid for each class. My vectors have some overlap, ultimately the goal is to draw a decision boundary through the data set for classification. I appreciate the link and it may answer my question, I think it's mostly my lack of understanding the material in general. – Logg Mar 25 '15 at 23:28
  • 1
    Welcome to StackOverflow! Your problem is pretty broad. This site can help you if you have a specific issue that you're struggling with, like your code not working as you expect or not being sure how to implement something specifically. Try to go as far as you can on your own and and come back if you have a specific question. You might try [this tutorial for knn](http://machinelearningmastery.com/tutorial-to-implement-k-nearest-neighbors-in-python-from-scratch/) first (it's in Python but you should be able to follow it). – eigenchris Mar 25 '15 at 23:31

0 Answers0