Add ExperimentConfig struct

.buildkite/pipeline.sh

@@ -28,7 +28,7 @@ steps:
   - label: Initialize
     key: init
     command: |
-      julia --project=$exp_dir -e 'import Pkg; Pkg.instantiate(;verbose=true)'
+      julia --project=$exp_dir -e 'import Pkg; Pkg.build("CalibrateAtmos"); Pkg.instantiate(;verbose=true)'
 EOM
 
 if [ "$generate_data" == "true" ] ; then

.github/workflows/ci.yml

@@ -39,7 +39,7 @@ jobs:
       - name: Surface Fluxes Perfect Model Test
         run: |
           julia --project=experiments/surface_fluxes_perfect_model -e 'using Pkg; Pkg.instantiate(;verbose=true)'
-          julia -t 10 --project=experiments/surface_fluxes_perfect_model test/test_sf.jl
+          julia --project=experiments/surface_fluxes_perfect_model test/test_sf.jl
       - uses: julia-actions/julia-runtest@v1
       - uses: julia-actions/julia-processcoverage@v1
       - uses: codecov/codecov-action@v3

Project.toml

@@ -10,9 +10,7 @@ ClimaParams = "5c42b081-d73a-476f-9059-fd94b934656c"
 Distributions = "31c24e10-a181-5473-b8eb-7969acd0382f"
 EnsembleKalmanProcesses = "aa8a2aa5-91d8-4396-bcef-d4f2ec43552d"
 JLD2 = "033835bb-8acc-5ee8-8aae-3f567f8a3819"
-PrecompileTools = "aea7be01-6a6a-4083-8856-8a6e6704d82a"
 Random = "9a3f8284-a2c9-5f02-9a11-845980a1fd5c"
-SciMLBase = "0bca4576-84f4-4d90-8ffe-ffa030f20462"
 TOML = "fa267f1f-6049-4f14-aa54-33bafae1ed76"
 YAML = "ddb6d928-2868-570f-bddf-ab3f9cf99eb6"
 
@@ -23,8 +21,6 @@ ClimaParams = "0.10"
 Distributions = "0.25"
 EnsembleKalmanProcesses = "1"
 JLD2 = "0.4"
-PrecompileTools = "1"
 Random = "1"
-SciMLBase = "1, 2"
 TOML = "1"
 YAML = "0.4"

docs/src/api.md

@@ -10,6 +10,12 @@ CalibrateAtmos.observation_map
 
 ## EnsembleKalmanProcesses Interface
 ```@docs
+CalibrateAtmos.calibrate
 CalibrateAtmos.initialize
+CalibrateAtmos.save_G_ensemble
 CalibrateAtmos.update_ensemble
+CalibrateAtmos.ExperimentConfig
+CalibrateAtmos.get_prior
+CalibrateAtmos.get_param_dict
+CalibrateAtmos.path_to_iteration
 ```

docs/src/experiment_setup_guide.md

@@ -148,8 +148,8 @@ Pseudocode for `observation_map(iteration)`:
 function observation_map(::Val{:sphere_held_suarez_rhoe_equilmoist}, iteration)
     # Get Configuration
     experiment_id = "sphere_held_suarez_rhoe_equilmoist"
-    config = load_config(experiment_id)
-    ensemble_size = config["ensemble_size"]
+    config = ExperimentConfig(experiment_id)
+    ensemble_size = config.ensemble_size
 
     # Setup output array
     # dims = size of individual member observation map output
@@ -215,17 +215,17 @@ The EKP configuration file must contain the following:
 - `n_iterations`, the number of iterations to run
 - `ensemble_size`, the ensemble size
 - `prior_path`, the path to the TOML file with the prior parameter distributions
-- `truth_data`, the truth data
-- `truth_noise`, the covariance of the truth data
+- `observations`, the observational data
+- `noise`, the covariance of the observational data
 - `output_dir`, the folder where you want calibration data and logs to be output. This must be the same as the `output_dir` in the model configuration file.
 Example:
 ```
 output_dir: output/sphere_held_suarez_rhoe_equilmoist
 prior_path: experiments/sphere_held_suarez_rhoe_equilmoist/prior.toml
 ensemble_size: 10
 n_iterations: 3
-truth_data: experiments/sphere_held_suarez_rhoe_equilmoist/obs_mean.jld2
-truth_noise: experiments/sphere_held_suarez_rhoe_equilmoist/obs_noise_cov.jld2
+observations: experiments/sphere_held_suarez_rhoe_equilmoist/obs_mean.jld2
+noise: experiments/sphere_held_suarez_rhoe_equilmoist/obs_noise_cov.jld2
 ```
 
 ## Plotting Results

docs/src/precompilation.md

@@ -16,11 +16,8 @@ import ClimaAtmos as CA
 import YAML
 
 @setup_workload begin
-    output = joinpath("precompilation")
     job_id = "your configuration"
-    config_file = joinpath("experiments", job_id, "atmos_config.yml")
-    config_dict = YAML.load_file(config_file)
-    config_dict["output_dir"] = output
+    ExperimentConfig(job_id; output_dir = "precompilation")
     @compile_workload begin
         initialize(job_id)
         config = CA.AtmosConfig(config_dict)

experiments/sphere_held_suarez_rhoe_equilmoist/ekp_config.yml

@@ -2,5 +2,5 @@ output_dir: output/sphere_held_suarez_rhoe_equilmoist
 prior_path: experiments/sphere_held_suarez_rhoe_equilmoist/prior.toml
 ensemble_size: 10
 n_iterations: 3
-truth_data: experiments/sphere_held_suarez_rhoe_equilmoist/obs_mean.jld2
-truth_noise: experiments/sphere_held_suarez_rhoe_equilmoist/obs_noise_cov.jld2
+observations: experiments/sphere_held_suarez_rhoe_equilmoist/obs_mean.jld2
+noise: experiments/sphere_held_suarez_rhoe_equilmoist/obs_noise_cov.jld2

experiments/sphere_held_suarez_rhoe_equilmoist/model_interface.jl

@@ -69,11 +69,7 @@ Runs the atmosphere model with the given an AtmosConfig object.
 Currently only has basic error handling.
 """
 
-function run_forward_model(
-    ::ClimaAtmosModel,
-    atmos_config::CA.AtmosConfig;
-    lk = nothing,
-)
+function run_forward_model(::ClimaAtmosModel, atmos_config::CA.AtmosConfig)
     simulation = CA.get_simulation(atmos_config)
     sol_res = CA.solve_atmos!(simulation)
     if sol_res.ret_code == :simulation_crashed

experiments/sphere_held_suarez_rhoe_equilmoist/observation_map.jl

@@ -3,15 +3,14 @@ using Statistics
 import YAML
 import EnsembleKalmanProcesses: TOMLInterface
 import JLD2
-import CalibrateAtmos: observation_map, get_ekp_config
+import CalibrateAtmos: observation_map, ExperimentConfig
 using ClimaAnalysis
 export observation_map
 
 function observation_map(::Val{:sphere_held_suarez_rhoe_equilmoist}, iteration)
     experiment_id = "sphere_held_suarez_rhoe_equilmoist"
-    config = get_ekp_config(experiment_id)
-    output_dir = config["output_dir"]
-    ensemble_size = config["ensemble_size"]
+    configuration = ExperimentConfig(experiment_id)
+    (; ensemble_size, output_dir) = configuration
 
     dims = 1
     G_ensemble = Array{Float64}(undef, dims..., ensemble_size)

experiments/surface_fluxes_perfect_model/Manifest.toml

@@ -4,11 +4,6 @@ julia_version = "1.10.2"
 manifest_format = "2.0"
 project_hash = "2edfa41926269520ea7e69208df499eef7522bd0"
 
-[[deps.ADTypes]]
-git-tree-sha1 = "41c37aa88889c171f1300ceac1313c06e891d245"
-uuid = "47edcb42-4c32-4615-8424-f2b9edc5f35b"
-version = "0.2.6"
-
 [[deps.AMD]]
 deps = ["LinearAlgebra", "SparseArrays", "SuiteSparse_jll"]
 git-tree-sha1 = "45a1272e3f809d36431e57ab22703c6896b8908f"
@@ -294,7 +289,7 @@ uuid = "83423d85-b0ee-5818-9007-b63ccbeb887a"
 version = "1.16.1+1"
 
 [[deps.CalibrateAtmos]]
-deps = ["CalibrateEmulateSample", "ClimaComms", "ClimaParams", "Distributions", "EnsembleKalmanProcesses", "JLD2", "PrecompileTools", "Random", "SciMLBase", "TOML", "YAML"]
+deps = ["CalibrateEmulateSample", "ClimaComms", "ClimaParams", "Distributions", "EnsembleKalmanProcesses", "JLD2", "Random", "TOML", "YAML"]
 path = "../.."
 uuid = "4347a170-ebd6-470c-89d3-5c705c0cacc2"
 version = "0.1.0"
@@ -396,11 +391,6 @@ git-tree-sha1 = "08c8b6831dc00bfea825826be0bc8336fc369860"
 uuid = "861a8166-3701-5b0c-9a16-15d98fcdc6aa"
 version = "1.0.2"
 
-[[deps.CommonSolve]]
-git-tree-sha1 = "0eee5eb66b1cf62cd6ad1b460238e60e4b09400c"
-uuid = "38540f10-b2f7-11e9-35d8-d573e4eb0ff2"
-version = "0.2.4"
-
 [[deps.CommonSubexpressions]]
 deps = ["MacroTools", "Test"]
 git-tree-sha1 = "7b8a93dba8af7e3b42fecabf646260105ac373f7"
@@ -769,17 +759,6 @@ git-tree-sha1 = "aa31987c2ba8704e23c6c8ba8a4f769d5d7e4f91"
 uuid = "559328eb-81f9-559d-9380-de523a88c83c"
 version = "1.0.10+0"
 
-[[deps.FunctionWrappers]]
-git-tree-sha1 = "d62485945ce5ae9c0c48f124a84998d755bae00e"
-uuid = "069b7b12-0de2-55c6-9aab-29f3d0a68a2e"
-version = "1.1.3"
-
-[[deps.FunctionWrappersWrappers]]
-deps = ["FunctionWrappers"]
-git-tree-sha1 = "b104d487b34566608f8b4e1c39fb0b10aa279ff8"
-uuid = "77dc65aa-8811-40c2-897b-53d922fa7daf"
-version = "0.1.3"
-
 [[deps.Future]]
 deps = ["Random"]
 uuid = "9fa8497b-333b-5362-9e8d-4d0656e87820"
@@ -1879,12 +1858,6 @@ git-tree-sha1 = "40b9edad2e5287e05bd413a38f61a8ff55b9557b"
 uuid = "5eaf0fd0-dfba-4ccb-bf02-d820a40db705"
 version = "0.2.1"
 
-[[deps.RuntimeGeneratedFunctions]]
-deps = ["ExprTools", "SHA", "Serialization"]
-git-tree-sha1 = "6aacc5eefe8415f47b3e34214c1d79d2674a0ba2"
-uuid = "7e49a35a-f44a-4d26-94aa-eba1b4ca6b47"
-version = "0.5.12"
-
 [[deps.SCS]]
 deps = ["MathOptInterface", "Requires", "SCS_GPU_jll", "SCS_jll", "SparseArrays"]
 git-tree-sha1 = "8d908b7c81e199ee92d17b6192849e8c43d2f31d"
@@ -1918,37 +1891,6 @@ git-tree-sha1 = "3aac6d68c5e57449f5b9b865c9ba50ac2970c4cf"
 uuid = "476501e8-09a2-5ece-8869-fb82de89a1fa"
 version = "0.6.42"
 
-[[deps.SciMLBase]]
-deps = ["ADTypes", "ArrayInterface", "CommonSolve", "ConstructionBase", "Distributed", "DocStringExtensions", "EnumX", "FunctionWrappersWrappers", "IteratorInterfaceExtensions", "LinearAlgebra", "Logging", "Markdown", "PrecompileTools", "Preferences", "Printf", "RecipesBase", "RecursiveArrayTools", "Reexport", "RuntimeGeneratedFunctions", "SciMLOperators", "StaticArraysCore", "Statistics", "SymbolicIndexingInterface", "Tables"]
-git-tree-sha1 = "3a281a9fce9cd62b849d7f16e412933a5fe755cb"
-uuid = "0bca4576-84f4-4d90-8ffe-ffa030f20462"
-version = "2.29.0"
-
-    [deps.SciMLBase.extensions]
-    SciMLBaseChainRulesCoreExt = "ChainRulesCore"
-    SciMLBaseMakieExt = "Makie"
-    SciMLBasePartialFunctionsExt = "PartialFunctions"
-    SciMLBasePyCallExt = "PyCall"
-    SciMLBasePythonCallExt = "PythonCall"
-    SciMLBaseRCallExt = "RCall"
-    SciMLBaseZygoteExt = "Zygote"
-
-    [deps.SciMLBase.weakdeps]
-    ChainRules = "082447d4-558c-5d27-93f4-14fc19e9eca2"
-    ChainRulesCore = "d360d2e6-b24c-11e9-a2a3-2a2ae2dbcce4"
-    Makie = "ee78f7c6-11fb-53f2-987a-cfe4a2b5a57a"
-    PartialFunctions = "570af359-4316-4cb7-8c74-252c00c2016b"
-    PyCall = "438e738f-606a-5dbb-bf0a-cddfbfd45ab0"
-    PythonCall = "6099a3de-0909-46bc-b1f4-468b9a2dfc0d"
-    RCall = "6f49c342-dc21-5d91-9882-a32aef131414"
-    Zygote = "e88e6eb3-aa80-5325-afca-941959d7151f"
-
-[[deps.SciMLOperators]]
-deps = ["ArrayInterface", "DocStringExtensions", "LinearAlgebra", "MacroTools", "Setfield", "SparseArrays", "StaticArraysCore"]
-git-tree-sha1 = "10499f619ef6e890f3f4a38914481cc868689cd5"
-uuid = "c0aeaf25-5076-4817-a8d5-81caf7dfa961"
-version = "0.3.8"
-
 [[deps.ScientificTypesBase]]
 git-tree-sha1 = "a8e18eb383b5ecf1b5e6fc237eb39255044fd92b"
 uuid = "30f210dd-8aff-4c5f-94ba-8e64358c1161"

experiments/surface_fluxes_perfect_model/ekp_config.yml

@@ -2,6 +2,6 @@ output_dir: output/surface_fluxes_perfect_model
 prior_path: experiments/surface_fluxes_perfect_model/prior.toml
 ensemble_size: 10
 n_iterations: 10
-truth_data: experiments/surface_fluxes_perfect_model/data/obs_mean.jld2
-truth_noise: experiments/surface_fluxes_perfect_model/data/obs_noise_cov.jld2
+observations: experiments/surface_fluxes_perfect_model/data/obs_mean.jld2
+noise: experiments/surface_fluxes_perfect_model/data/obs_noise_cov.jld2
 generate_data: true

experiments/surface_fluxes_perfect_model/model_interface.jl

@@ -1,7 +1,11 @@
 import EnsembleKalmanProcesses as EKP
 using CalibrateAtmos
 import CalibrateAtmos:
-    AbstractPhysicalModel, get_config, run_forward_model, get_forward_model
+    AbstractPhysicalModel,
+    get_config,
+    run_forward_model,
+    get_forward_model,
+    ExperimentConfig
 import YAML
 
 """
@@ -43,13 +47,12 @@ function get_config(
     iteration,
     experiment_id::AbstractString,
 )
-    config_dict = YAML.load_file("experiments/$experiment_id/model_config.yml")
-    return get_config(model, member, iteration, config_dict)
+    return get_config(model, member, iteration, ExperimentConfig(experiment_id))
 end
 
 """
     get_config(member, iteration, experiment_id::AbstractString)
-    get_config(member, iteration, config_dict::AbstractDict)
+    get_config(member, iteration, experiment_config::ExperimentConfig)
 
 Returns an config dictionary object for the given member and iteration.
 If given an experiment id string, it will load the config from the corresponding YAML file.
@@ -60,22 +63,29 @@ function get_config(
     ::SurfaceFluxModel,
     member,
     iteration,
-    config_dict::AbstractDict,
+    experiment_config::ExperimentConfig,
 )
     # Specify member path for output_dir
-    output_dir = config_dict["output_dir"]
+    model_config = YAML.load_file(
+        joinpath(
+            "experiments",
+            "surface_fluxes_perfect_model",
+            "model_config.yml",
+        ),
+    )
+    output_dir = (experiment_config.output_dir)
     # Set TOML to use EKP parameter(s)
     member_path =
         EKP.TOMLInterface.path_to_ensemble_member(output_dir, iteration, member)
-    config_dict["output_dir"] = member_path
+    model_config["output_dir"] = member_path
     parameter_path = joinpath(member_path, "parameters.toml")
-    if haskey(config_dict, "toml")
-        push!(config_dict["toml"], parameter_path)
+    if haskey(model_config, "toml")
+        push!(model_config["toml"], parameter_path)
     else
-        config_dict["toml"] = [parameter_path]
+        model_config["toml"] = [parameter_path]
     end
 
-    return config_dict
+    return model_config
 end
 
 """
@@ -84,18 +94,8 @@ end
 Runs the model with the given an AbstractDict object.
 """
 
-function run_forward_model(
-    ::SurfaceFluxModel,
-    config::AbstractDict;
-    lk = nothing,
-)
-    x_inputs = if isnothing(lk)
-        load_profiles(config["x_data_file"])
-    else
-        lock(lk) do
-            load_profiles(config["x_data_file"])
-        end
-    end
+function run_forward_model(::SurfaceFluxModel, config::AbstractDict)
+    x_inputs = load_profiles(config["x_data_file"])
     FT = typeof(x_inputs.profiles_int[1].T)
     obtain_ustar(FT, x_inputs, config)
 end

experiments/surface_fluxes_perfect_model/observation_map.jl

@@ -2,7 +2,7 @@ using Statistics
 import YAML
 import EnsembleKalmanProcesses: TOMLInterface
 import JLD2
-import CalibrateAtmos: observation_map, get_ekp_config
+import CalibrateAtmos: observation_map
 
 """
     observation_map(::Val{:surface_fluxes_perfect_model}, iteration)
@@ -11,10 +11,8 @@ Returns the observation map (from the raw model output to the observable y),
 as specified by process_member_data, for the given iteration.
 """
 function observation_map(::Val{:surface_fluxes_perfect_model}, iteration)
-    experiment_id = "surface_fluxes_perfect_model"
-    config = get_ekp_config(experiment_id)
-    output_dir = config["output_dir"]
-    ensemble_size = config["ensemble_size"]
+    config = ExperimentConfig("surface_fluxes_perfect_model")
+    (; output_dir, ensemble_size) = config
     model_output = "model_ustar_array.jld2"
 
     dims = 1

experiments/surface_fluxes_perfect_model/plot.jl

@@ -104,20 +104,17 @@ pkg_dir = pkgdir(CalibrateAtmos)
 model_config = YAML.load_file(
     joinpath(pkg_dir, "experiments", experiment_id, "model_config.yml"),
 )
-ekp_config = YAML.load_file(
-    joinpath(pkg_dir, "experiments", experiment_id, "ekp_config.yml"),
-)
-N_iter = ekp_config["n_iterations"]
-N_mem = ekp_config["ensemble_size"]
+ekp_config = ExperimentConfig(experiment_id)
+N_iter = ekp_config.n_iterations
+N_mem = ekp_config.ensemble_size
 eki_path = joinpath(
     joinpath(pkg_dir, model_config["output_dir"]),
     "iteration_$(lpad(N_iter, 3, '0'))",
     "eki_file.jld2",
 );
 eki = JLD2.load_object(eki_path);
 EKP.get_u(eki)
-prior_path = joinpath(pkg_dir, ekp_config["prior_path"])
-prior = CalibrateAtmos.get_prior(prior_path)
+prior = ekp_config.prior
 
 theta_star_vec =
     (; coefficient_a_m_businger = 4.7, coefficient_a_h_businger = 4.7)