spectralpython / spectral

Python module for hyperspectral image processing
MIT License
571 stars 139 forks source link

Add denoising methods to class PrincipalComponents #91

Closed gemmaellen closed 5 years ago

gemmaellen commented 5 years ago

Hi there!

I wrote a couple of denoising methods into the PrincipalComponents class, with a format similar to the ones in the MNFResult class. I'm currently using MNF on some data I'm working on, and I find it useful to be able to swiftly compare the MNF denoising to an equivalent PCA denoising.

This is my first ever pull request, just so you know, so I hope I'm going about this in basically the right way. Tips are welcome.

Thanks! --Gemma

gemmaellen commented 5 years ago

It looks like the checks aren't failing because of anything I did -- I've run them on an unaltered branch and they fail there, too, in the exact same way. I think the issue is this one:

https://travis-ci.community/t/issue-with-python-2-6-on-linux/3861

The suggested fix of adding dist: trusty to the travis.yml file works. I'm not sure if this is the actual fix you wish to implement, or if you want to no longer support Python 2.6 and Python 3.3, instead.

tboggs commented 5 years ago

Hi Gemma,

Thanks for the addition! I'll accept the merge request shortly

Also, I'm long overdue for pushing out an updated release so I'll try to get that done within the next week or so.

Thanks, Thomas