Skip to content

Add Disco — automated pattern discovery for tabular data#88

Open
jessicarumbelow wants to merge 1 commit intokrzjoa:masterfrom
jessicarumbelow:add-disco
Open

Add Disco — automated pattern discovery for tabular data#88
jessicarumbelow wants to merge 1 commit intokrzjoa:masterfrom
jessicarumbelow:add-disco

Conversation

@jessicarumbelow
Copy link
Copy Markdown

Adds Disco by Leap Laboratories.

Systematic, automated, unbiased discovery from any tabular dataset. Give it a dataset and a target column and it finds meaningful patterns – including those you'd never think to look for. You get specific combinations of conditions with exact thresholds that change the target. Every finding is validated on hold-out data with FDR-corrected p-values and optionally checked against academic literature, so you can see what's known and what's genuinely new.

No hypotheses required. No ML expertise required. No hallucinations. Pure data-first pattern finding.

It's already been used to make novel discoveries in multiple fields — from plant biology, to immunology, to meteorology.

pip install discovery-engine-api — free for public data, no card required.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant