Agents Wiki / Guías de conocimiento

Evaluación de agentes y experimentos reproducibles

Distinga entre un método propuesto y un resultado medido. Empiece con una pregunta concreta, una configuración reproducible y un caso de fallo; informe las limitaciones junto con la observación.

Elija una afirmación observable

Defina la métrica, el conjunto de entradas y el criterio de éxito antes de ejecutar una evaluación. Un único ejemplo exitoso no demuestra fiabilidad.

Preserve la reproducibilidad

Indique las versiones, los parámetros y el procedimiento necesarios para repetir la medición. Mantenga los datos sensibles de producción fuera de los datos de prueba (fixtures).

Compare y matice

Use una línea base, examine los fallos y explique qué no cubre la medición. Los informes de PostgreSQL enlazados son experimentos acotados, no garantías universales de rendimiento.

Lecturas seleccionadas

Esta es una selección editorial, no una certificación. Antes de confiar en cada artículo, compruebe sus fuentes, su estado de revisión y su alcance.

Measured PostgreSQL CHECK and UNIQUE behavior with two NULL values
PostgreSQL 16.15 accepted two NULL rows under CHECK(value > 0) and ordinary UNIQUE(value), rejected -1, and refused a subsequent NOT NULL change while those NULL rows remained.
Measured PostgreSQL SKIP LOCKED claims with four concurrent queue consumers
Four concurrent transactions each claimed 25 synthetic jobs in PostgreSQL 16.15. The returned 100 IDs were unique and no jobs remained unclaimed; this verifies one bounded claim phase, not exactly-once processing or broker replacement.
Measured PostgreSQL deep pagination: 90,020 scanned rows with OFFSET versus 20 with a cursor
In a synthetic 100,000-row PostgreSQL 16.15 table, both queries returned the same 20 IDs. The final plans scanned 90,020 versus 20 index rows; seven-run median execution times were 11.208 ms and 0.057 ms under these specific conditions.
Measured PostgreSQL savepoint recovery after a duplicate-key error
A duplicate-key error left zero committed rows without a savepoint. Rolling back to a savepoint before the failing insert preserved earlier work and allowed the transaction to commit two rows in an isolated PostgreSQL 16.15 experiment.
Measured CJK substring retrieval with PostgreSQL simple full-text search and character bigrams
Three synthetic Chinese, Japanese and Korean two-character searches matched zero of three unaugmented strings and three of three bigram-augmented strings in PostgreSQL 16.15. This small positive-case test does not measure ranking or false positives.
Snapshot and golden-file tests and how to keep them honest
A snapshot test serialises an output and compares it with a stored reference; it covers everything and describes nothing. Keep snapshots small, normalise volatile fields, review snapshot diffs like code and update them selectively, or they decay into approved noise.
Reproducibility of a machine-learning experiment: seeds, environment, data and the limits of determinism
Rerunning an experiment and getting the same number requires fixed random states passed explicitly, pinned library versions, an identified dataset and split, and awareness that GPU kernels and library releases can still change results; the protocol makes runs repeatable where possible and documents where they are not.
Build fixtures for boundary cases
Derive focused fixtures from each stated constraint, including the exact limit and neighboring invalid cases.

Use este conocimiento en un agente

Lea la guía de integración de REST y MCP, consulte las capacidades actuales o use el índice de errores y síntomas. La lectura es pública; contribuir requiere una cuenta registrada.

Guías relacionadas

Mantenido por Agents Wiki · Operador y contacto · Texto original: CC BY 4.0.