Merge RT/Engines, LUT structure#953
Conversation
| pairs = zip(self.lut_names, x_RT) | ||
| return " ".join([f"{name}={val:5.3f}" for name, val in pairs]) | ||
|
|
||
| # REVIEW: We need to think about the best place for the two albedo method (here, radiative_transfer.py, utils, etc.) |
There was a problem hiding this comment.
would motion to remove this review note, and that I think it sounded last week like this atmosphere.py was the best place.
| # convert observer zenith to MODTRAN convention if needed | ||
| if self.indices.convert_observer_zenith: | ||
| point[self.indices.convert_observer_zenith] = ( | ||
| 180.0 - point[self.indices.convert_observer_zenith] | ||
| ) | ||
|
|
There was a problem hiding this comment.
Double check this should/does still exist for the Modtran case? Also, was looking for it because this would be a nice one to add to units.py too I think.
|
Huge progress! Posted small patch PR here from items initially caught: https://github.com/evan-greenbrg/isofit/compare/refactor/rte...pgbrodrick:isofit:rte_patches?expand=1 (sorry, forgot to suppress black). Small changes worth double checking, but I think they will help. Done prior to 13f9eba. |
|
More work to get the small cube example to run to completion (yay) in evan-greenbrg#7 @brentwilder , can you review? Particularly the question on coszen in fileio (will examine with fresh eyes tomorrow). Evan noted he's okay with direct pushes, here, but I'd prefer some quick checks on these first. Now merged. |
| # coszen stored as scalar in the LUT | ||
| self.coszen = float(forward.atmosphere.lut["coszen"].values) |
There was a problem hiding this comment.
This similar line needs to be included in analytical_line.py and algebraic_line.py
| self.glint_model = engine_config.glint_model | ||
| if self.glint_model and not self.multipart_transmittance: | ||
| raise AttributeError( | ||
| "Using the glint model requires a multipart transmittance LUT table" |
There was a problem hiding this comment.
with the config, glint_model gone in this update, do we need a different guardrail to prevent user from running glint without multipart? Or perhaps this removed because we already check this somewhere else in the code?
|
More test help on evan-greenbrg#8 - if we're okay with the backwards compatibility check, this should be good to go. |
dc5214e to
c35892a
Compare
load_esd in analytical line
Black Circular Import Circular import Circular imports Test debugging
Resample debugging Resample issues Resampling order
c35892a to
4a53edb
Compare
I pushed up draft of the LUT-atmosphere architecture. I drew up an accompanying diagram.
In short:
loading.
That's the idea anyway. Still need to work through debugging.
Status:
Cleanup and testing