Retrieve an ExoTransmit forward model + noise about platon

[ideasrule]

This is what I get when I do that on PyExoTransmit-generated WFC3 and Spitzer data. The lines on the corner plots show the true values. These were all made with retrieve_fake_data.py, which I've committed to the repository.

The retrieval doesn't do badly. I've tried running the retrieval more times, and there are times when the retrieved values are 2-3 sigma off. I don't know how concerning these are--we retrieve 5 parameters per run and I've done ~30 runs, so just by random chance some outliers are expected.

10 ppm error on transit depths:

50 ppm error:

100 ppm error:

1000 ppm error:

[ideasrule]

Finally, here's an example of what you get with ExoTransmit (not PyExoTransmit) with 50 ppm error. I've ran it 3 times; the other two times, the results were substantially worse than the attached (~2 sigma off), then a little bit better.