Edit on GitHub

Model Search

MLJ has a model registry, allowing the user to search models and their properties, without loading all the packages containing model code. In turn, this allows one to efficiently find all models solving a given machine learning task. The task itself is specified with the help of the matching method, and the search executed with the models methods, as detailed below.

Model metadata

Terminology. In this section the word "model" refers to a metadata entry in the model registry, as opposed to an actual model struct that such an entry represents. One can obtain such an entry with the info command:

julia> info("PCA")
Principal component analysis. Learns a linear transformation to project the data  on a lower dimensional space while preserving most of the initial variance.
→ based on [MultivariateStats](https://github.com/JuliaStats/MultivariateStats.jl).
→ do `@load PCA pkg="MultivariateStats"` to use the model.
→ do `?PCA` for documentation.
(name = "PCA",
 package_name = "MultivariateStats",
 is_supervised = false,
 docstring = "Principal component analysis. Learns a linear transformation to project the data  on a lower dimensional space while preserving most of the initial variance.\n→ based on [MultivariateStats](https://github.com/JuliaStats/MultivariateStats.jl).\n→ do `@load PCA pkg=\"MultivariateStats\"` to use the model.\n→ do `?PCA` for documentation.",
 hyperparameter_ranges = (nothing, nothing, nothing, nothing),
 hyperparameter_types = ("Union{Nothing, Int64}", "Symbol", "Float64", "Union{Nothing, Array{Float64,1}, Real}"),
 hyperparameters = (:maxoutdim, :method, :pratio, :mean),
 implemented_methods = Symbol[:clean!, :fit, :fitted_params, :transform],
 is_pure_julia = true,
 is_wrapper = false,
 load_path = "MLJModels.MultivariateStats_.PCA",
 package_license = "MIT",
 package_url = "https://github.com/JuliaStats/MultivariateStats.jl",
 package_uuid = "6f286f6a-111f-5878-ab1e-185364afe411",
 prediction_type = :unknown,
 supports_online = false,
 supports_weights = false,
 input_scitype = Table{_s23} where _s23<:(AbstractArray{_s25,1} where _s25<:Continuous),
 target_scitype = Table{_s23} where _s23<:(AbstractArray{_s25,1} where _s25<:Continuous),
 output_scitype = Unknown,)

So a "model" in the present context is just a named tuple containing metadata, and not an actual model type or instance. If two models with the same name occur in different packages, the package name must be specified, as in info("LinearRegressor", pkg="GLM").

General model queries

We list all models (named tuples) using models(), and list the models for which code is already loaded with localmodels():

julia> localmodels()
10-element Array{NamedTuple{(:name, :package_name, :is_supervised, :docstring, :hyperparameter_ranges, :hyperparameter_types, :hyperparameters, :implemented_methods, :is_pure_julia, :is_wrapper, :load_path, :package_license, :package_url, :package_uuid, :prediction_type, :supports_online, :supports_weights, :input_scitype, :target_scitype, :output_scitype),T} where T<:Tuple,1}:
 (name = ConstantClassifier, package_name = MLJModels, ... )
 (name = ConstantRegressor, package_name = MLJModels, ... )
 (name = ContinuousEncoder, package_name = MLJModels, ... )
 (name = FeatureSelector, package_name = MLJModels, ... )
 (name = FillImputer, package_name = MLJModels, ... )
 (name = OneHotEncoder, package_name = MLJModels, ... )
 (name = Standardizer, package_name = MLJModels, ... )
 (name = UnivariateBoxCoxTransformer, package_name = MLJModels, ... )
 (name = UnivariateDiscretizer, package_name = MLJModels, ... )
 (name = UnivariateStandardizer, package_name = MLJModels, ... )

julia> localmodels()[2]
Constant regressor (Probabilistic).
→ based on [MLJModels](https://github.com/alan-turing-institute/MLJModels.jl).
→ do `@load ConstantRegressor pkg="MLJModels"` to use the model.
→ do `?ConstantRegressor` for documentation.
(name = "ConstantRegressor",
 package_name = "MLJModels",
 is_supervised = true,
 docstring = "Constant regressor (Probabilistic).\n→ based on [MLJModels](https://github.com/alan-turing-institute/MLJModels.jl).\n→ do `@load ConstantRegressor pkg=\"MLJModels\"` to use the model.\n→ do `?ConstantRegressor` for documentation.",
 hyperparameter_ranges = (nothing,),
 hyperparameter_types = ("Type{D} where D<:Distributions.Sampleable",),
 hyperparameters = (:distribution_type,),
 implemented_methods = Symbol[:fitted_params, :predict],
 is_pure_julia = true,
 is_wrapper = false,
 load_path = "MLJModels.ConstantRegressor",
 package_license = "MIT",
 package_url = "https://github.com/alan-turing-institute/MLJModels.jl",
 package_uuid = "d491faf4-2d78-11e9-2867-c94bc002c0b7",
 prediction_type = :probabilistic,
 supports_online = false,
 supports_weights = false,
 input_scitype = Table,
 target_scitype = AbstractArray{Continuous,1},
 output_scitype = Unknown,)

One can search for models containing specified strings or regular expressions in their docstring attributes, as in

julia> models("forest")
4-element Array{NamedTuple{(:name, :package_name, :is_supervised, :docstring, :hyperparameter_ranges, :hyperparameter_types, :hyperparameters, :implemented_methods, :is_pure_julia, :is_wrapper, :load_path, :package_license, :package_url, :package_uuid, :prediction_type, :supports_online, :supports_weights, :input_scitype, :target_scitype, :output_scitype),T} where T<:Tuple,1}:
 (name = RandomForestClassifier, package_name = DecisionTree, ... )
 (name = RandomForestClassifier, package_name = ScikitLearn, ... )
 (name = RandomForestRegressor, package_name = DecisionTree, ... )
 (name = RandomForestRegressor, package_name = ScikitLearn, ... )

or by specifying a filter (Bool-valued function):

julia> filter(model) = model.is_supervised &&
                       model.input_scitype >: MLJ.Table(Continuous) &&
                       model.target_scitype >: AbstractVector{<:Multiclass{3}} &&
                       model.prediction_type == :deterministic
filter (generic function with 1 method)

julia> models(filter)
12-element Array{NamedTuple{(:name, :package_name, :is_supervised, :docstring, :hyperparameter_ranges, :hyperparameter_types, :hyperparameters, :implemented_methods, :is_pure_julia, :is_wrapper, :load_path, :package_license, :package_url, :package_uuid, :prediction_type, :supports_online, :supports_weights, :input_scitype, :target_scitype, :output_scitype),T} where T<:Tuple,1}:
 (name = DeterministicConstantClassifier, package_name = MLJModels, ... )
 (name = LinearSVC, package_name = LIBSVM, ... )
 (name = NuSVC, package_name = LIBSVM, ... )
 (name = PassiveAggressiveClassifier, package_name = ScikitLearn, ... )
 (name = PerceptronClassifier, package_name = ScikitLearn, ... )
 (name = RidgeCVClassifier, package_name = ScikitLearn, ... )
 (name = RidgeClassifier, package_name = ScikitLearn, ... )
 (name = SGDClassifier, package_name = ScikitLearn, ... )
 (name = SVC, package_name = LIBSVM, ... )
 (name = SVMClassifier, package_name = ScikitLearn, ... )
 (name = SVMLinearClassifier, package_name = ScikitLearn, ... )
 (name = SVMNuClassifier, package_name = ScikitLearn, ... )

Multiple test arguments may be passed to models, which are applied conjunctively.

Matching models to data

Common searches are streamlined with the help of the matching command, defined as follows:

  • matching(model, X, y) == true exactly when model is supervised and admits inputs and targets with the scientific types of X and y, respectively

  • matching(model, X) == true exactly when model is unsupervised and admits inputs with the scientific types of X.

So, to search for all supervised probabilistic models handling input X and target y, one can define the testing function task by

task(model) = matching(model, X, y) && model.prediction_type == :probabilistic

And execute the search with

models(task)

Also defined are Bool-valued callable objects matching(model), matching(X, y) and matching(X), with obvious behaviour. For example, matching(X, y)(model) = matching(model, X, y).

So, to search for all models compatible with input X and target y, for example, one executes

models(matching(X, y))

while the preceding search can also be written

models() do model
    matching(model, X, y) &&
    model.prediction_type == :probabilistic
end

API

MLJBase.models — Function
models()

List all models in the MLJ registry. Here and below model means the registry metadata entry for a genuine model type (a proxy for types whose defining code may not be loaded).

models(conditions...)

List all models satisifying the specified conditions. A condition is any Bool-valued function on models.

Excluded in the listings are the built-in model-wraps EnsembleModel, TunedModel, and IteratedModel.

Example

If

task(model) = model.is_supervised && model.is_probabilistic

then models(task) lists all supervised models making probabilistic predictions.

See also: localmodels.

models(needle::Union{AbstractString,Regex})

List all models whole name or docstring matches a given needle.

models(N::AbstractNode)

A vector of all models referenced by a node N, each model appearing exactly once.

MLJModels.localmodels — Function
localmodels(; modl=Main)
localmodels(conditions...; modl=Main)
localmodels(needle::Union{AbstractString,Regex}; modl=Main)

List all models whose names are in the namespace of the specified module modl, or meeting the conditions, if specified. Here a condition is a Bool-valued function on models.

See also models

MLJ.matching — Function

matching(model, X, y)

Returns true exactly when the registry metadata entry model is supervised and admits inputs and targets with the scientific types of X and y, respectively.

matching(model, X)

Returns true exactly when model is unsupervised and admits inputs with the scientific types of X.

matching(model), matching(X, y), matching(X)

Curried versions of the preceding methods, i.e., Bool-valued callable objects satisfying matching(X, y)(model) = matching(model, X, y), etc.

Example

models(matching(X))

Finds all unsupervised models compatible with input data X.

models() do model
    matching(model, X, y) && model.prediction_type == :probabilistic
end

Finds all supervised models compatible with input data X and target data y and making probabilistic predictions.

See also models

source