raddleverse / MimiCIAM.jl

MIT License
2 stars 1 forks source link

WIP: Added protect-only model and tests for comment #1

Closed cledna closed 6 years ago

cledna commented 6 years ago

@davidanthoff, @FrankErrickson, @lrennels
This is still very much a work in progress but I wanted to put it up here so everyone could comment. I did a mock-up of a protect-only case evaluating a single segment. The main problems it still has are as follows:

  1. I have no idea why it gives negative values for the second adaptation period; that's a bug I'm working on tracking down.
  2. Data integration; country to segment and segment to country mapping
  3. I'm not quite clear on the deal with lslr and lslrPlan and their difference; that's something I want to ask Delavane. I just made something up for testing purposes
  4. Comparison with GAMS results
  5. Test cases that look at different choices of first and final adaptation periods (for evaluation of model robustness)

I've also flagged some things in the code that I'll open separate issues for. Overall, I'm hoping to get some feedback on how I structured the code (too many if-statements?) and ways to streamline it and make it more efficient. Looking forward to your thoughts!