selecting rows in numpy ndarray based on the value of two columns

Question

I have a big np.ndarray (3600000,3), the HUE, the VALUE, and an associated CLASS number. For each pairs of HUE a

Answer 1

Try this code:

 x[x[:, 2] == class_number[:, :2]

where x is np.ndarray

 x[:, 2] == class_number

contains true/false that means whether the last is class_number or not.

You need to take a look at: Boolean indexing in http://wiki.scipy.org/Cookbook/Indexing

Moved from comment.