Anthony Blaom, PhD

Checking model instances are not types

1

A very common gotcha for users (not necessarily beginners) is calling a method with a model *type* instead of a model *instance*. Sometimes the consequence is a confusing error message....

Julia crashes for multithreaded Stack for some non-Julia models

4

Context: #767 adds support for an option `acceleration=CPUThreads()` in composite model types defined by "exporting" learning networks, and implements this option for `Stack`. I have been carrying out [MLJ ecosystem...

Needs: - [x] Resolution of https://github.com/JuliaData/Tables.jl/pull/278 Alternatively, if https://github.com/JuliaML/MLUtils.jl/issues/61 can be resolved, we go straight to `getobs`.

Opening this issue after a nice suggestion of @davnn . Some clusterers (eg, sckitlearn's DBSCAN) only deliver labels for the training data and cannot immediately label new unseen data. In...

`serializable` does not remove data implicitly stored in partial functions, leading to scaling with data size and potential privacy breaches

6

**edit:** New issue title more accurately explains the issue not immediately diagnosed in the original comment below. --- This issue is not related to the master or dev branches but...

This trait is defined but not currently used anywhere in MLJ (and not part of the public API). No model overloads it, but it has a fallback using `methodswith(typeof(model))`. Similar...

Disintegration of MLJBase (discussion and tracking issue)

12

It seems there is a case to be made for further modularisation of MLJBase. These all sound reasonable to me but happy to hear other ideas: - [x] Separate MLJSerialization...

Add scitype checks to predict, transform, and so forth

1

In src/operations.jl