setup yaml (#231)

* setup yaml

* get settings

* write and read work now

Author: lazarusA

Project.toml

@@ -23,6 +23,7 @@ MLUtils = "f1d291b0-491e-4a28-83b9-f70985020b54"
 NCDatasets = "85f8d34a-cbdd-5861-8df4-14fed0d494ab"
 NamedDims = "356022a1-0364-5f58-8944-0da4b18d706f"
 OptimizationOptimisers = "42dfb2eb-d2b4-4451-abcd-913932933ac1"
+OrderedCollections = "bac558e1-5e72-5ebc-8fee-abe8a469f55d"
 PrettyTables = "08abe8d2-0d0c-5749-adfa-8a2ac140af0d"
 Printf = "de0858da-6303-5e67-8744-51eddeeeb8d7"
 ProgressMeter = "92933f4c-e287-5a05-a399-4b506db050ca"
@@ -31,6 +32,7 @@ Reexport = "189a3867-3050-52da-a836-e630ba90ab69"
 Static = "aedffcd0-7271-4cad-89d0-dc628f76c6d3"
 Statistics = "10745b16-79ce-11e8-11f9-7d13ad32a3b2"
 StyledStrings = "f489334b-da3d-4c2e-b8f0-e476e12c162b"
+YAML = "ddb6d928-2868-570f-bddf-ab3f9cf99eb6"
 Zygote = "e88e6eb3-aa80-5325-afca-941959d7151f"
 
 [weakdeps]
@@ -60,12 +62,14 @@ Makie = "0.22, 0.23, 0.24"
 NCDatasets = "0.14.8"
 NamedDims = "1.2.3"
 OptimizationOptimisers = "0.3.7"
+OrderedCollections = "1.8.1"
 PrettyTables = "2.4.0, 3.1.2"
 ProgressMeter = "1.10.4"
 Reexport = "1.2.2"
 Static = "1.3.1"
 Statistics = "1"
 StyledStrings = "1.0.3, 1.11.0"
+YAML = "0.4.16"
 Zygote = "0.7.7"
 julia = "1.10"
 

src/EasyHybrid.jl

@@ -26,13 +26,15 @@ using MLJ: partition
 using MLUtils: MLUtils, DataLoader, kfolds, numobs, rpad, splitobs
 using NCDatasets: NCDatasets, NCDataset, close, name
 using OptimizationOptimisers: OptimizationOptimisers, AdamW, Adam, Optimisers
+using OrderedCollections: OrderedDict
 using PrettyTables: PrettyTables
 using Printf: Printf, @sprintf
 using ProgressMeter: ProgressMeter, Progress, next!
 using Random: Random, AbstractRNG, randperm, randstring
 using Reexport: @reexport
 using Statistics: Statistics, mean, cor, quantile, var
 using StyledStrings: StyledStrings, @styled_str
+using YAML: load_file, write_file
 using Zygote: Zygote
 using Static: False, True
 
@@ -66,5 +68,6 @@ include("utils/helpers_for_HybridModel.jl")
 include("utils/helpers_data_loading.jl")
 include("tune.jl")
 include("utils/helpers_cross_validation.jl")
+include("utils/config_yaml.jl")
 
 end

src/train.jl

@@ -354,6 +354,14 @@ function train(
     train_diffs = !isempty(set_diff) ? NamedTuple{Tuple(set_diff)}([getproperty(ŷ_train, e) for e in set_diff]) : nothing
     val_diffs = !isempty(set_diff) ? NamedTuple{Tuple(set_diff)}([getproperty(ŷ_val, e) for e in set_diff]) : nothing
 
+    # collect all training arguments into a NamedTuple for easier saving and logging
+    train_args = (; nepochs, batchsize, opt, patience, autodiff_backend, return_gradients, array_type, training_loss, loss_types, extra_loss, agg, train_from, random_seed, file_name, hybrid_name, return_model, monitor_names, folder_to_save, plotting, show_progress, yscale)
+
+    # get config settings and save to yaml file
+    get_config_settings = get_full_config(hybridModel, train_args)
+    path_yaml = joinpath(tmp_folder, "config_settings.yaml")
+    save_hybrid_config(get_config_settings, path_yaml)
+
     # TODO: save/output metrics
     return TrainResults(
         train_history,
@@ -366,7 +374,7 @@ function train(
         ps,
         st,
         best_epoch,
-        best_agg_loss
+        best_agg_loss,
     )
 end
 

src/utils/config_yaml.jl

@@ -0,0 +1,33 @@
+export load_hybrid_config, save_hybrid_config
+export get_hybrid_config, get_train_config, get_full_config
+
+function load_hybrid_config(path::String; dicttype = OrderedDict{String, Any})
+    return load_file(path; dicttype)
+end
+
+function save_hybrid_config(config::OrderedDict, path::String)
+    return write_file(path, config)
+end
+
+function get_hybrid_config(hm::LuxCore.AbstractLuxContainerLayer)
+    hm_config = OrderedDict{String, Any}()
+    for field in fieldnames(typeof(hm))
+        hm_config[string(field)] = getfield(hm, field)
+    end
+    return hm_config
+end
+
+function get_train_config(train_args::NamedTuple)
+    train_config = OrderedDict{String, Any}()
+    for field in fieldnames(typeof(train_args))
+        train_config[string(field)] = getfield(train_args, field)
+    end
+    return train_config
+end
+
+function get_full_config(hm::LuxCore.AbstractLuxContainerLayer, train_args::NamedTuple)
+    full_config = OrderedDict{String, Any}()
+    full_config["hybrid_model"] = get_hybrid_config(hm)
+    full_config["train_args"] = get_train_config(train_args)
+    return full_config
+end