Skip to content

Latest commit

 

History

History
118 lines (80 loc) · 4.6 KB

ROADMAP.md

File metadata and controls

118 lines (80 loc) · 4.6 KB
Raw

Road map

February 2020

Please visit contributing guidelines if interested in contributing to MLJ.

Guiding goals

  • Usability, interoperability, extensibility, reproducibility, and code transparency.

  • Offer state-of-art tools for model composition and model optimization (hyper-parameter tuning)

  • Avoid common pain-points of other frameworks:

    • identifying all models that solve a given task

    • routine operations requiring a lot of code

    • passage from data source to algorithm-specific data format

    • probabilistic predictions: inconsistent representations, lack of options for performance evaluation

  • Add some focus to julia machine learning software development more generally

Priorities

Priorities are somewhat fluid, depending on funding offers and available talent. Rough priorities for the core development team at present are marked with † below. However, we are always keen to review external contributions in any area.

Future enhancements

Adding models

Enhancing core functionality

  • † Iterative model control #139

  • † Add more tuning strategies. HyperOpt.jl integration. Particular focus on random search, Bayesian methods, and AD-powered gradient descent. See here for complete wish-list. #74 #38 #37

  • Systematic benchmarking, probably modeled on MLaut #74

  • Give EnsembleModel more extendible API and extend beyond bagging (boosting, etc) and migrate to separate repository? #363

  • † Enhance complex model compostition, in particular stacking (POC) #311 #282

Broadening scope

  • Spin-off a stand-alone measures (loss functions) package (currently here)

  • Add sparse data support (NLP); could use NaiveBayes.jl as test case (currently wrapped only for dense input)

  • POC for implementation of time series models #303, ScientificTypes #14

  • Add tools or separate repository for visualization in MLJ. Only end-to-end visualization provided now is for two-parameter model tuning #85 (closed) #416 #342

  • Add more pre-processing tools, enhance MLJScientificType's autotype method.

Scalability

  • Online learning support and distributed data #60

  • DAG scheduling for learning network training #72 (multithreading first?)

  • Automated estimates of cpu/memory requirements #71

  • Add multithreading to tuning MLJTuning #15