Cosmetic issue with graphing TI in cases with zeros for some lambdas

[davidlmobley]

There is a cosmetic issue with graphing the dU/dlambda (dH/dlambda) curve in some cases with some lambda vectors being unchanging at zero, wherein there's a blank region in the graph. The data linked below will yield one such example.

I've also observed in some other cases, there are only x axis labels for some lambda components and not others.

Data for the first case: https://dl.dropboxusercontent.com/u/3409095/guilherme.zip

[dbonanni]

Hi David, I would to thank you and the others developers for alchemical analysis, such a good tool that is helping me a lot. I think I'm finding the same issue. I am not able to see your pictures but here is the link of some graphing I have obtained by alchemical analysis (https://drive.google.com/open?id=0BxeyUUtR8RLOWGFQZzkzOTdzVTA). It's not very important issue but it would be great fix the problem, do you have a solution?

Cheers,

Davide Bonanni

[davidlmobley]

Thanks, @dbonanni . I know what the issue is (basically, the code makes too many assumptions about what lambda pathways you might be following and how to split those into components) but fixing it will require rather significant changes. As I don't currently have anyone actively developing this code, I'm waiting on: a) alchemlyb to become mature enough (or maybe it is already) that it can replace much of the analysis code here and we can rewrite this mostly to focus on assessment of results/plotting/etc. rather than focusing as much on parsing, and b) someone suitable who is going to start developing this again in my group

Thanks!

[dbonanni]

Thank you so much for your prompt reply @davidlmobley. I will look forward to future developments.

Good job and have a great day!