search

jamesrobertlloyd / gpss-research

Kernel structure discovery research code - likely to be unstable
MIT License
187 stars 73 forks source link

Only show posterior variance #57

Open jamesrobertlloyd opened 10 years ago

jamesrobertlloyd commented 10 years ago

Without the posterior mean - might be less misleading?

duvenaud commented 10 years ago

That's a great idea. Although, there's also nothing special about +- 2SD either - what if we plotted posterior density continuously using shading or color?

duvenaud commented 10 years ago

Something like this: sweet