[Sage Discussion]: Pure/Mixture data split

openforcefield / nistdataselection

Records the tools and decisions used to select NIST data for curation.

MIT License

3 stars 0 forks source link

[Sage Discussion]: Pure/Mixture data split #17

Open ocmadin opened 4 years ago

ocmadin commented 4 years ago

Pure/Mixture data split: In fitting, how should we prioritize pure vs. mixture data? In each set, what do we emphasize to cover the deficiencies of the other sets?

davidlmobley commented 4 years ago

Are you asking how many datapoints of each kind?

I don't think we know this. I don't have an intuition. What are you, @mrshirts @leeping and @SimonBoothroyd thinking?

SimonBoothroyd commented 4 years ago

I also don't have a particularly good intuition for this, and I think our plan is to (time permitting) do some initial exploratory studies to this end.