Datalad generator
Model Selection Code Generator
Compare candidate models on identical folds, tune the one worth tuning, and get a performance figure you can actually quote, because the test set never took part in choosing it.
1
Your data
Read in your browser, never uploaded. Best run on the output of the feature engineering generator, but any modelling table will do.
Drop a CSV here, or click to choose one
Large files are sampled for profiling. Nothing leaves this page.
Share and embed
Use this tool on your own site
Free to embed, no attribution required, though a credit link is appreciated and included by default. The tool runs entirely in the visitor's browser, so nothing you enter is sent anywhere.
px
[…] standard error rule, favouring the simpler model Holdout scored once, at the end, not while tuning Open the generator Free · runs in your browser · no […]
[…] One standard error rule, so noise does not win Pick the target, the metric and the fold count Open the generator Free · runs in your browser · no […]
[…] One standard error rule, favouring the simpler model Choose the target, the metric and the folds Open the generator Free · runs in your browser · no […]
[…] One standard error rule, so noise does not decide Pick the target, the metric and the fold count Open the generator Free · runs in your browser · no […]
[…] standard error rule, so a small gain does not win Pick the target, the metric and the fold count Open the generator Free · runs in your browser · no […]