microprediction
@microprediction
Public Skills
precise
by microprediction
Online (incremental) covariance, correlation, and precision estimation in Python — the streaming complement to sklearn.covariance. Use when code needs a covariance/correlation matrix updated per observation, recomputes np.cov/np.corrcoef in a rolling loop, must judge or compare covariance estimates, or proposes a new covariance methodology. Points to task-specific skills.
HumpDay code-review skill
by microprediction
Repository: <https://github.com/microprediction/humpday>
skaters code-review skill
by microprediction
uses skaters appropriately, say so and move on — don't manufacture offences.
conformal-coverage-check
by microprediction
Guard rails for conformal prediction. Use when a user writes, fits, calibrates, reviews, or chooses between conformal / split-conformal models (conformalguide, MAPIE, crepes, puncc, TorchCP, or hand-rolled), reports or targets a coverage level, or asks whether their coverage is reliable, conditional, per-case, or trustworthy. Two jobs: review the code for common misuses of conformal prediction, and run the conformalguide.guardrails distance-covariance check for conditionally uneven coverage.