34 lines (22 loc) · 1011 Bytes

hyperloglog

Language: Python · Sphere: programming · Category: Performance

Signature: () → None

What it does

HyperLogLog probabilistic cardinality estimator implementation.

This module provides a HyperLogLog class for estimating the cardinality (number of distinct elements) of large datasets with minimal memory usage.

Guarantee

When it runs, hyperloglog guarantees abs(est - 1000.0) < 50.0; abs(est2 - 1000.0) < 50.0; abs(merged - 1500.0) < 75.0 (proven by run).

Checkable constraints:

abs(est - 1000.0) < 50.0
est_after == est
abs(est2 - 1000.0) < 50.0
abs(merged - 1500.0) < 75.0
hll.estimate_cardinality() == merged

Verification evidence

Green-run: ✓ passes (re-run under the extractor's gate)
Constraint strength: recovery (truth-pinned)
Independent oracle: — none yet (green-run candidate; not an axiom under the frozen ruler)
Peer review: unreviewed

△ AURA Pattern Library — © Reality Optimizer