Surf/activation example with very different name

[trontrytel]

This PR adds an example to the documentation. The example shows how to calibrate the AerosolActivation module from CloudMicrophysics.jl package. The calibration is done in perfect model setting.

[odunbar]

I get an error after i instantiate:

    Status `~/Dropbox/Caltech/CESjl/EnsembleKalmanProcesses.jl/examples/AerosolActivation/Project.toml`
  [6eacf6c3] CLIMAParameters v0.2.0
  [6a9e3e04] CloudMicrophysics v0.3.0
  [31c24e10] Distributions v0.24.18
  [aa8a2aa5] EnsembleKalmanProcesses v0.1.0
  [91a5bcdd] Plots v1.22.1
  [37e2e46d] LinearAlgebra
  [44cfe95a] Pkg
  [9a3f8284] Random

and then call the example out of the box:

ERROR: LoadError: MethodError: no method matching total_N_activated(::EarthParameterSet, ::CloudMicrophysics.AerosolModel.AerosolDistribution{Tuple{CloudMicrophysics.AerosolModel.Mode_B{Tuple{Float64}, Float64}}}, ::Float64, ::Float64, ::Float64)
Closest candidates are:
  total_N_activated(::CLIMAParameters.AbstractParameterSet, ::CloudMicrophysics.AerosolModel.AbstractAerosolDistribution, ::FT, ::FT, ::FT, ::Thermodynamics.PhasePartition{FT}) where FT<:Real at /home/odunbar/.julia/packages/CloudMicrophysics/gL0Zn/src/AerosolActivation.jl:274
Stacktrace:
 [1] run_activation_model(molar_mass_calibrated::Float64, osmotic_coeff_calibrated::Float64)
   @ Main ~/Dropbox/Caltech/CESjl/EnsembleKalmanProcesses.jl/examples/AerosolActivation/aerosol_activation.jl:125
 [2] top-level scope
   @ ~/Dropbox/Caltech/CESjl/EnsembleKalmanProcesses.jl/examples/AerosolActivation/aerosol_activation.jl:139
in expression starting at /home/odunbar/Dropbox/Caltech/CESjl/EnsembleKalmanProcesses.jl/examples/AerosolActivation/aerosol_activation.jl:139

[trontrytel]

I get an error after i instantiate:

    Status `~/Dropbox/Caltech/CESjl/EnsembleKalmanProcesses.jl/examples/AerosolActivation/Project.toml`
  [6eacf6c3] CLIMAParameters v0.2.0
  [6a9e3e04] CloudMicrophysics v0.3.0
  [31c24e10] Distributions v0.24.18
  [aa8a2aa5] EnsembleKalmanProcesses v0.1.0
  [91a5bcdd] Plots v1.22.1
  [37e2e46d] LinearAlgebra
  [44cfe95a] Pkg
  [9a3f8284] Random

and then call the example out of the box:

ERROR: LoadError: MethodError: no method matching total_N_activated(::EarthParameterSet, ::CloudMicrophysics.AerosolModel.AerosolDistribution{Tuple{CloudMicrophysics.AerosolModel.Mode_B{Tuple{Float64}, Float64}}}, ::Float64, ::Float64, ::Float64)
Closest candidates are:
  total_N_activated(::CLIMAParameters.AbstractParameterSet, ::CloudMicrophysics.AerosolModel.AbstractAerosolDistribution, ::FT, ::FT, ::FT, ::Thermodynamics.PhasePartition{FT}) where FT<:Real at /home/odunbar/.julia/packages/CloudMicrophysics/gL0Zn/src/AerosolActivation.jl:274
Stacktrace:
 [1] run_activation_model(molar_mass_calibrated::Float64, osmotic_coeff_calibrated::Float64)
   @ Main ~/Dropbox/Caltech/CESjl/EnsembleKalmanProcesses.jl/examples/AerosolActivation/aerosol_activation.jl:125
 [2] top-level scope
   @ ~/Dropbox/Caltech/CESjl/EnsembleKalmanProcesses.jl/examples/AerosolActivation/aerosol_activation.jl:139
in expression starting at /home/odunbar/Dropbox/Caltech/CESjl/EnsembleKalmanProcesses.jl/examples/AerosolActivation/aerosol_activation.jl:139

Right... I did release a new CloudMicrophysics.jl version this morning. Lemme fix the deps

[trontrytel]

I updated the version of CloudMicrophysics.jl

@odunbar - do you know where to put the additional dependencies for the documenter build to be aware of them?

[odunbar]

@trontrytel have you tried adding them to the docs/Project.toml, and then add using statements in the docs/Make.jl? I believe this would be natural as it is the documenter that has to run the example file. It only doesnt need to add EnsembleKalmanProcesses.jl there because it instead locally adds this in Make.jl with the filepath

@navidcy wrote some nice literate docs for LossMinimization, and perhaps can concur?

[navidcy]

Sure -- I can help out with that! :)
(after the Ocean Meeting Abstract deadline in a few hours :))

[navidcy]

hey sorry, this slipped my radar! I'll get to it now!

[navidcy]

A remark: when the docs are being build the use the docs/Project.toml file to load deps. If an example requires some deps outside of the EnsempleKalmanProcesses.jl package itself then they should be include in the docs/Project.toml. I added those deps (see 71fe662) and now docs build should be able to go through and produce a docs preview at:

https://clima.github.io/EnsembleKalmanProcesses.jl/previews/PR54/literated/aerosol_activation/

[navidcy]

I made some comments. Seems like the example is still a work in progress?

[navidcy]

Some more comments.

[trontrytel]

A remark: when the docs are being build the use the docs/Project.toml file to load deps. If an example requires some deps outside of the EnsempleKalmanProcesses.jl package itself then they should be include in the docs/Project.toml. I added those deps (see 71fe662) and now docs build should be able to go through and produce a docs preview at:

https://clima.github.io/EnsembleKalmanProcesses.jl/previews/PR54/literated/aerosol_activation/

Thank you for taking a look and fixing the dependencies! It's great to see the example building online. I'll go through your review comments right now and finish it.

[trontrytel]

@navidcy - Thank you for all the comments and making it work! I applied the formatter and fixed a typo in plot name. So I hope that all four plots will show up now and that the formatter won't complain any more.

Can I squash/rebase and force push here to get rid of the cluttered history?

[navidcy]

@navidcy - Thank you for all the comments and making it work! I applied the formatter and fixed a typo in plot name. So I hope that all four plots will show up now and that the formatter won't complain any more.

Can I squash/rebase and force push here to get rid of the cluttered history?

No worries!

Sure, do whatever you like! :)

[navidcy]

@trontrytel perhaps edit the first comment of the PR and possibly also its title to make it a bit more descriptive? It sort of reads a bit esoteric now.

[trontrytel]

@trontrytel perhaps edit the first comment of the PR and possibly also its title to make it a bit more descriptive? It sort of reads a bit esoteric now.

Right... I just noticed that too :) No Jupyter notebook survived the process. Will update

[navidcy]

I added a few more suggestions regarding some typos, some simplifications to Julia code, adding markers in the ensemble mean plots, and rounding the estimated values.

Nice example!

[navidcy]

@trontrytel how do you run the formatter offline? Apparently some of the changes I suggested were not inline with the formatter's style.

[trontrytel]

@trontrytel how do you run the formatter offline? Apparently some of the changes I suggested were not inline with the formatter's style.

fixed!

[navidcy]

@trontrytel how do you run the formatter offline? Apparently some of the changes I suggested were not inline with the formatter's style.

fixed!

Nice!!

But still, if you know how to run the formatter offline can you tell me?

[trontrytel]

@trontrytel how do you run the formatter offline? Apparently some of the changes I suggested were not inline with the formatter's style.

fixed!

Nice!!

But still, if you know how to run the formatter offline can you tell me?

Sure!

julia --project=.dev/ .dev/climaformat.jl file_to_be_formatted.jl

I'm superstitious so I run it on individual files that I know I changed. But you can look into the logs of the JuliaFormatter run and in the first command line of Apply JuliaFormatter he does the same but for all julia --project=.dev .dev/climaformat.jl .

[trontrytel]

bors r+

[trontrytel]

bors r-

[bors]

Canceled.

[trontrytel]

bors r+

[bors]

Build succeeded:

• docs-build
• format
• test_success