Egutierrez e142ef026d fix(eda): hallazgos de comportamiento del benchmark (H2,H3,H6,H7,H8,H10,H11) ...

Ronda 4 (verificada con re-corrida sobre los datasets afectados):
- H2: stl_decompose deriva periodo de la frecuencia del indice (seattle period=365
  seasonal_strength=0.84; fin del period=2 espurio)
- H3+H10: infer_fk por senal de nombre (<X>Id->X.<X>Id) + excluir no-clave -> chinook
  111->9 FK, todas reales, cero absurdas, 16-27x mas rapido; base intacta (flag off->111)
- H6: association no computa eta2 si cardinalidad~=n (Ticket-Fare espurio fuera)
- H7: id secuencial monotono excluido de correlacion y PCA/KMeans (PassengerId fuera)
- H8: correlacion de series no estacionarias marcada espuria / sobre retornos
- H11: distribution_type usa modos/cardinalidad/normalidad (quality->discrete)
- 66 tests verdes

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

2026-06-29 06:37:47 +02:00

..

__init__.py

test(datascience): corpus stub para gliner_load_model + extract_entities_gliner

2026-04-30 16:33:46 +02:00

test_align_relations_to_entities.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_alpha_shape_concave_hull.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_best_central_tendency.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_detect_distribution_type.py

fix(eda): hallazgos de comportamiento del benchmark (H2,H3,H6,H7,H8,H10,H11)

2026-06-29 06:37:47 +02:00

test_extract_entities_gliner.py

test(datascience): corpus stub para gliner_load_model + extract_entities_gliner

2026-04-30 16:33:46 +02:00

test_extract_graph_gliner2.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_extract_relations_glirel.py

test(datascience): corpus stub para glirel_load_model + extract_relations_glirel

2026-04-30 16:41:14 +02:00

test_extract_relations_mrebel.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_extract_triples_spacy_es.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_fuzzy_merge_adaptive.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_geometric_mean.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_gliner2_load_model.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_kde_density_levels.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_parse_rebel_output.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_plot_heatmap_log.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_plot_kde_2d.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_remove_words_from_column.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_spacy_es_load_model.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_summary_stats.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_translate_es_to_en.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_trimmed_mean.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00

test_vault_csv_profile.py

chore: auto-commit (95 archivos)

2026-05-13 00:50:34 +02:00

test_vault_pdf_extract.py

chore: auto-commit (95 archivos)

2026-05-13 00:50:34 +02:00

test_words_to_dataset.py

feat: extraccion masiva footprint_aurgi (41 funcs + 4 types + stack Docker geo)

2026-05-04 23:35:22 +02:00