search

Anmol6 / DNGO-BO

Bayesian optimization with DNGO (Deep Networks for Global Optimization)
11 stars 4 forks source link

Nicer plots showing the GP's marginal distribution #2

Open mgelbart opened 8 years ago

mgelbart commented 8 years ago

http://matplotlib.org/api/pyplot_api.html#matplotlib.pyplot.fill_between

Anmol6 commented 7 years ago

Yup, using this

mgelbart commented 7 years ago

Can you please indicate where in this repo one can find this plot, and/or paste the plot in this thread?

mgelbart commented 7 years ago

We looked at

image

You should use fill_between and make improvements in #7. Then we need to figure out why the dark and light blue don't intersect at the data points and why the uncertainty is constant everywhere rather than being smaller near the data. You should also explicitly show the data as points on the graph (can use '.' in plot).

mgelbart commented 7 years ago

I think this bug was fixed but we still should switch to using fill_between for aesthetic reasons.