Raporlama

EarthquakeAPI.run_sync() ve run_async() basarili oldugunda PipelineResult dondurur.

result = api.run_sync(criteria, strategy.get_name())
pipeline_result = result.value

PipelineResult alanlari

Alan	Tip	Aciklama
selected_df	pd.DataFrame	Secilen kayitlar
scored_df	pd.DataFrame	Tum puanlanan kayitlar
report	dict	JSON'a yazilabilir rapor
execution_time	float	Calisma suresi
failed_providers	list[str]	Veri cekemeyen provider'lar
logs	list[str]	Pipeline olaylari

DataFrame izlenebilirlik kolonlari

scored_df tum kayitlar icin su kolonlari tasir:

• SCORE
• SCORE_BREAKDOWN
• SELECTION_STATUS
• SELECTION_REASON
• EVENT_GROUP: ayni depremin farkli provider'lardaki adlarini birlestiren olay kimligi (tek provider'da <PROVIDER>|<EVENT>); "ayni depremden en fazla 3 kayit" siniri bu kolona gore uygulanir.

SELECTION_STATUS:

• selected
• rejected

SELECTION_REASON ornekleri:

• selected
• score_below_min_score:55.0
• max_per_station:3
• max_per_event:3
• num_records_limit:11

report alanlari

report = pipeline_result.report

Temel alanlar:

• status
• search_criteria
• selected_count
• total_considered
• strategy
• providers
• records
• statistics

Izlenebilirlik alanlari:

• selection_summary
• score_breakdown
• deduplication: birlestirilen olay gruplari ve cikarilan tekrar kayit sayisi
• compliance: secilen/istenen kayit sayisi, eksik kayit (shortfall), ayni depremden secilen en fazla kayit ve max_per_event siniri, warnings, compliant (TBDY: her yon 11 kayit => 22, ayni depremden en fazla 3)

Selection summary

print(report["selection_summary"])

Ornek:

{
    "status_counts": {"selected": 11, "rejected": 42},
    "rejection_reasons": {
        "score_below_min_score:55.0": 24,
        "max_per_event:3": 10,
        "num_records_limit:11": 8,
    },
}

JSON rapor yazma

import json

with open("selection_report.json", "w", encoding="utf-8") as fp:
    json.dump(pipeline_result.report, fp, indent=2, ensure_ascii=False, default=str)