From 736b9b3e319ca85ef30c467122705804c7379324 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?P=C3=A1ll=20Haraldsson?= Date: Tue, 17 Jan 2023 11:10:12 +0000 Subject: [PATCH 1/3] Typos [skip ci] --- src/accessor_functions.jl | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/src/accessor_functions.jl b/src/accessor_functions.jl index 8746c44..e3e142b 100644 --- a/src/accessor_functions.jl +++ b/src/accessor_functions.jl @@ -8,18 +8,18 @@ const ACCESSOR_FUNCTIONS = ( """ LearnAPI.feature_importances(model, fitted_params, report) -Return the model-specific feature importances of `model`, given `fittted_params` and +Return the model-specific feature importances of `model`, given `fitted_params` and `report`, as returned by [`LearnAPI.fit`](@ref), [`LearnAPI.update!`](@ref) or [`LearnAPI.ingest!`](@ref). The value returned has the form of an abstract vector of -`feature::Symbol => importance::Real` pairs (e.g `[:gender =>0.23, :height =>0.7, :weight +`feature::Symbol => importance::Real` pairs (e.g `[:gender => 0.23, :height => 0.7, :weight => 0.1]`). The `model` supports feature importances if `:feature_importance in LearnAPI.functions(model)`. If for some reason a model is sometimes unable to report feature importances, then -`feature_importances` will return all importances as 0.0, as in `[:gender =>0.0, :height -=>0.0, :weight => 0.0]`. +`feature_importances` will return all importances as 0.0, as in `[:gender => 0.0, :height +=> 0.0, :weight => 0.0]`. # New model implementations @@ -36,7 +36,7 @@ function feature_importances end """ training_losses(model, fitted_params, report) -Return the training losses for `model`, given `fittted_params` and +Return the training losses for `model`, given `fitted_params` and `report`, as returned by [`LearnAPI.fit`](@ref), [`LearnAPI.update!`](@ref) or [`LearnAPI.ingest!`](@ref). @@ -53,7 +53,7 @@ function training_losses end """ training_scores(model, fitted_params, report) -Return the training scores for `model`, given `fittted_params` and +Return the training scores for `model`, given `fitted_params` and `report`, as returned by [`LearnAPI.fit`](@ref), [`LearnAPI.update!`](@ref) or [`LearnAPI.ingest!`](@ref). @@ -71,7 +71,7 @@ function training_scores end """ training_labels(model, fitted_params, report) -Return the training labels for `model`, given `fittted_params` and +Return the training labels for `model`, given `fitted_params` and `report`, as returned by [`LearnAPI.fit`](@ref), [`LearnAPI.update!`](@ref) or [`LearnAPI.ingest!`](@ref). From 4884bfe186bdc0f6db3cb892f186a242358c7ca1 Mon Sep 17 00:00:00 2001 From: "Anthony D. Blaom" Date: Mon, 23 Jan 2023 12:13:57 +1300 Subject: [PATCH 2/3] clarify meaning of `ingest!` --- docs/src/index.md | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git a/docs/src/index.md b/docs/src/index.md index 9bd5a21..95edd32 100644 --- a/docs/src/index.md +++ b/docs/src/index.md @@ -58,7 +58,8 @@ The following methods, dispatched on model type, are provided: - `update!`, for adding model iterations, or responding efficiently to other post-`fit`changes in hyperparameters -- `ingest!`, for incremental learning +- `ingest!`, for incremental learning (training further using *new* data, without + re-initializing learned parameters) - **operations**, `predict`, `predict_joint`, `transform` and `inverse_transform` for applying the model to data not used for training From e170e2e3975c3e23e9f9f723d69819ea85f6b383 Mon Sep 17 00:00:00 2001 From: "Anthony D. Blaom" Date: Tue, 24 Jan 2023 13:45:55 +1300 Subject: [PATCH 3/3] add "Goals" section to landing page of docs --- docs/src/index.md | 20 ++++++++++++++++++-- 1 file changed, 18 insertions(+), 2 deletions(-) diff --git a/docs/src/index.md b/docs/src/index.md index 95edd32..27af1dc 100644 --- a/docs/src/index.md +++ b/docs/src/index.md @@ -4,13 +4,29 @@ LearnAPI.jl
-A basic Julia interface for training and applying machine learning models +A basement-level Julia interface for training and applying machine learning models

``` +## Goals + +- Ease of implementation for existing machine learning algorithms + +- Applicability to a large variety of algorithms + +- Provision of clear interface points for model-generic tooling, such as performance + evaluation through resampling, hyperparameter optimization, and iterative model control. + +- Should be data container agnostic + +- Should be documented in detail + +It is *not* a design goal of LearnAPI.jl to provide a convenient interface for the general +user to directly interact with ML models. + ## Quick tours -- To see how to **USE** models implementing LearnAPI: [Basic fit/predict +- To see how to **INTERACT WITH** models implementing LearnAPI: [Basic fit/predict workflow](@ref workflow). - For developers wanting to **IMPLEMENT** LearnAPI: [Anatomy of