Univariate GP - suggested code change
Thanks for a fantastic package!
I was playing around with it and found the following issue - I hope this is helpful to report.
I was trying to modify example_training_gp_pymelon.py to use a univariate input model.
I needed to modify two lines of code to make this work: In utils.py:
- L57 : replace ".squeeze" with ".flatten" so length scale parameters remain array-like
- L108 : replace ".squeeze" with ".flatten" so length scale parameters remain array-like
The original code for the 2D problem remains functional with the change. I would suggest to make the above changes in the code.