Closed bas-rustenburg closed 9 years ago
Diff looks kinda messy, probably want to look at the file on its own.
Edit: link now points to latest version in PR branch.
Here is something interesting.
I took a look at the different step methods for MCMC. Here is the host concentration, estimated at 1.0157 mM in this particular experiment (I just picked a host-guest titration at random from one of our previous sets of experiments).
Settings: Using the TwoComponentBindingModel about 320000 iterations (I stopped it at around 16% of 2000000) 10000 burn in iterations and a thinning period of 25. Not sure how good those settings are but there is quite a striking difference.
Using the RescalingStep: Note that the value on the y-axis is offset by 1.0124 Using a default pymc Metropolis step:
Opened an issue for discussion.
Good catch. Let's use MetropolisStep for now. I think eventually we want to use both!
Looking good, I like the base classes
@jchodera, could you take a quick look at this before I merge?
Looks great!
I was playing around with some of the code in TwoComponentBindingModel and thinking of ways to improve modularity. I wonder if this is more, or less readable than the current form.