DataScienceTutorials.jl

DataScienceTutorials.jl copied to clipboard

Proposal to redesign the landing page

10

I find the landing page for Data Science Tutorials extremely busy. The current organization around the four basic categories in the right hand menu ("Data Basics", "Getting Started", and so...

ablaom

## tutorials * [x] remove explicit import for auc etc #21 * [ ] add tutorial for StaticTransformer/FeatureSelector #24 * [ ] add bootstrapping to ISL5 #22 * [ ]...

tlienart

tlienart

Add the MNIST example from MLJFlux/examples

2

This includes a demonstration of wrapping a model in `IteratedModel` which would be new here.

ablaom

End-to-end examples -- UCI datasets

1

Can reproduce / take inspiration from [this repo](https://github.com/rupakc/UCI-Data-Analysis) which does a similar thing with Scikitlearn * [x] wine dataset * [x] horse colic (has missings) [here](https://archive.ics.uci.edu/ml/datasets/Horse+Colic) * [ ] abalone...

tlienart

Tuning at end of King County example does not improve score

1

https://alan-turing-institute.github.io/DataScienceTutorials.jl/end-to-end/HouseKingCounty/ cc @vollmersj

ablaom

tlienart

Training a bunch of models and comparing

3

This post: https://discourse.julialang.org/t/custom-xgboost-loss-function-w-zygote-julia-computing-blog-post/35811/13?u=tlienart by @azev77

tlienart

 The comment there does not seem to apply (since I don't know when). Perhaps the parameter could be reset to an optimal...

ablaom

Create Tutorial 3Models.ipynb by Cruz

7

A new tutorial showcasing MLJ by C. Cruz

drcxcruz