-1

I am trying to model some 1D data where each data point has a different error bar.

In the GPFlow documentation I find an example of how to do this here (known noise variances demo)

However, the demo doesn't seem to produce a GP model that models the data well and I was wondering if anyone had any suggests to improve the GP fit in the demo?

cyberface
  • 91
  • 5
  • Can you be a bit more specific about your problem and what is not working for you? Posting a snippet to reproduce the issue would help a lot. Thanks – Vincent Dutordoir Jul 09 '20 at 08:34
  • 1
    This was actually a bug in the notebook example, which is fixed [by this PR](https://github.com/GPflow/GPflow/pull/1526). This question itself should probably be closed/deleted, or if you still have issues please clarify your question! – STJ Jul 09 '20 at 09:41
  • Thanks for the replies and thanks for the PR! I will close this for now and give the PR ago when it's merged. – cyberface Jul 10 '20 at 10:32
  • 1
    @cyberface is this ready to close? the PR is merged – joel Jul 20 '20 at 21:23
  • @joel yeah it's working! Thank you so much! – cyberface Jul 22 '20 at 09:08

1 Answers1

0

This was fixed in this PR. It was a typo in the documentation.

cyberface
  • 91
  • 5