P3 - Lambda Testing #321
P3 - Lambda Testing #321
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #321 +/- ##
==========================================
+ Coverage 98.71% 98.72% +0.01%
==========================================
Files 33 33
Lines 1088 1101 +13
==========================================
+ Hits 1074 1087 +13
Misses 14 14 ☔ View full report in Codecov by Sentry. |
p3_sandbox/p3_lambdaTests.jl
Outdated
| import RootSolvers as RS | ||
| import CairoMakie as Plt | ||
|
|
||
| FT = Float64 |
There was a problem hiding this comment.
Do things also work for Float32 now, or is that still an issue given our units?
There was a problem hiding this comment.
That is still an issue
p3_sandbox/p3_lambdaTests.jl
Outdated
| ) where {FT} | ||
|
|
||
| ρ_rs = range(ρ_r_min, stop = ρ_r_max, length = numPlots) | ||
|
|
There was a problem hiding this comment.
Based on the plots shown here, could we add more test cases to the p3 scheme tests? Maybe not 100x100 for each ρ_rs but 10x10. We could just check if the error is lower than some eps estimated from the plots?
p3_sandbox/p3_lambdaTests.jl
Outdated
| λSteps::Int, | ||
| F_rSteps::Int, | ||
| ) where {FT} | ||
| λs = range(λ_min, stop = λ_max, length = λSteps) |
There was a problem hiding this comment.
If you are testing for Float32 and running into types issues it may be that we need FT(λ_min) here
| @@ -228,7 +228,7 @@ Returns the shape parameter μ for a given λ value | |||
| Eq. 3 in Morrison and Milbrandt (2015). | |||
| """ | |||
| function DSD_μ(p3::PSP3, λ::FT) where {FT} | |||
| @assert λ > FT(0) | |||
| #@assert λ > FT(0) | |||
There was a problem hiding this comment.
Should we have an assert like this somewhere in the P3 scheme code? (Maybe not necessarily here)
There was a problem hiding this comment.
Lets try with the gamma functions
anastasia-popova
force-pushed
the
ap/lambdas
branch
8 times, most recently
from
19c96c8 to
e966b7b
Compare
anastasia-popova
force-pushed
the
ap/lambdas
branch
from
8532881 to
bdaa5d6
Compare
Purpose
Find range of Lambdas for which the shape solver is successful
To-do
Content
This PR contains the updated lambda linear guesser, and updated docs with plots
Still TO DO in following PR: make the assert statement in mu function work (potentially through taking the max between a value and the calculated lambda so it is never negative) and add more extensive tests for lambda calculator