In D-Se/ML: Machine Leanring targets-bookdown pipeline

Compendium

Goals

• Deliver a scalable, reproducible, persistent and continuously integrated data analysis compendium that:

  • Deploys Machine Learning algorithms,

  • Visualizes model performances & statistical inference,

  • Communicates findings to target audiences in multiple formats.

• Display best practices of machine learning using R, concretely:

  • Packaging approaches using version control,

  • Deploying computationally efficient algorithms,

  • Interfacing modelling & reporting API's,

  • Adopting a functional programming mindset.

the Pipeline

click and drag on the image to inspect or move nodes!

### TODO: annotate visnetwork with time of last model run
targets::tar_visnetwork(label = "time")

::: note

When a file in the pipeline is edited, the dependencies get rerun and results are updated! :::

An example of such dependency is this parameterized book. If new data is added, the next time the pipeline is run, it gets cleaned, the relevant models get rerun and the book chapters get updated.

D-Se/ML documentation built on April 1, 2022, 10:53 p.m.