Frequently Asked Questions

Julia already has a great machine learning toolbox, ScitkitLearn.jl. Why MLJ?

An alternative machine learning toolbox for Julia users is ScikitLearn.jl. Initially intended as a Julia wrapper for the popular python library scikit-learn, ML algorithms written in Julia can also implement the ScikitLearn.jl API. Meta-algorithms (systematic tuning, pipelining, etc) remain python wrapped code, however.

While ScitkiLearn.jl provides the Julia user with access to a mature and large library of machine learning models, the scikit-learn API on which it is modeled, dating back to 2007, is not likely to evolve significantly in the future. MLJ enjoys (or will enjoy) several features that should make it an attractive alternative in the longer term:

• One language. ScikitLearn.jl wraps python code, which in turn wraps C code for performance-critical routines. A Julia machine learning algorithm that implements the MLJ model interface is 100% Julia. Writing code in Julia is almost as fast as python and well-written Julia code runs almost as fast as C. Additionally, a single language design provides superior interoperability. For example, one can implement: (i) gradient-descent tuning of hyperparameters, using automatic differentiation libraries such as Flux.jl; and (ii) GPU performance boosts without major code refactoring, using CuArrays.jl.

• Registry for model metadata. In ScikitLearn.jl the list of available models, as well as model metadata (whether a model handles categorical inputs, whether is can make probabilistic predictions, etc) must be gleaned from documentation. In MLJ, this information is more structured and is accessible to MLJ via an external model registry (without the models needing to be loaded). This forms the basis of a "task" interface and facilitates model composition (see below).

• Task interface. Once the MLJ user specifies a "task" (e.g., "make probabilistic predictions of home value, based on features x, y, z") then MLJ can automatically search for models matching that task, assisting in systematic benchmarking and model selection.

• Flexible API for model composition. Pipelines in scikit-learn are more of an afterthought than an integral part of the original design. By contrast, MLJ's user-interaction API was predicated on the requirements of a flexible "learning network" API, one that allows models to be connected in essentially arbitrary ways (including target transforming and inverse-transforming). Networks can be built and tested in stages before being exported as first-class stand-alone models. Networks feature "smart" training (only necessary components are retrained after parameter changes) and will eventually be trainable using a DAG scheduler. With the help of Julia's meta-programming features, constructing common architectures, such as linear pipelines and stacks, will be one-line operations.

• Clean probabilistic API. The scikit-learn API does not specify a universal standard for the form of probabilistic predictions. By fixing a probabilistic API along the lines of the skpro project, MLJ aims to improve support for Bayesian statistics and probabilistic graphical models.

• Universal adoption of categorical data types. Python's scientific array library NumPy has no dedicated data type for representing categorical data (i.e., no type that tracks the pool of all possible classes). Generally scikit-learn models deal with this by requiring data to be relabeled as integers. However, the naive user trains a model on relabeled categorical data only to discover that evaluation on a test set crashes his code because a categorical feature takes on a value not observed in training. MLJ mitigates such issues by insisting on the use of categorical data types, and by insisting that MLJ model implementations preserve the class pools. If, for example, a training target contains classes in the pool that do not actually appear in the training set, a probabilistic prediction will nevertheless predict a distribution whose support includes the missing class, but which is appropriately weighted with probability zero.

Finally, we note that there is a project underway to implement (some of) the ScikitLearn.jl models as MLJ models, as an temporary expedient.