# Taking Fitting Seriously

**URL:** <https://discourse.julialang.org/t/taking-fitting-seriously/13281>\
**Category:** Data\
**Tags:** plotting\
**Created:** [August 12, 2018, 8:50am UTC](https://discourse.julialang.org/t/taking-fitting-seriously/13281 "2018-08-12T08:50:21Z")\
**Posts on this page:** 1\
**Showing post:** 37

<div class="post-metadata">

**Author:** ![simonbyrne](https://sea2.discourse-cdn.com/julialang/user_avatar/discourse.julialang.org/simonbyrne/32/19_2.png) [@simonbyrne](https://discourse.julialang.org/u/simonbyrne)\
**Post date:** [September 6, 2018, 6:54pm UTC](https://discourse.julialang.org/t/taking-fitting-seriously/13281/37 "2018-09-06T18:54:39Z")

</div>

Another thing that came up on Slack is [GitHub - ablaom/Koala.jl: Julia machine learning environment](https://github.com/ablaom/Koala.jl), which appears to be it’s own little (unregistered) ML ecosystem. It has a few interesting ideas, one being that data splitting (e.g. train/test, or cross-validation) is just a wrapper type around the full dataset, along with a binary mask.

---

_[View the full topic](https://discourse.julialang.org/t/taking-fitting-seriously/13281)._