WHAT IT ANSWERS
这张表适合回答什么
- 查看某机制腿是否通过校准
- 比较 lift 与 FDR
- 审计 sealed run 结论
DATASET · research
research.daybreak_calibration_verdict
按 sealed run 和机制腿家族记录 lift、FDR、置信区间及校准结论,用于判断信号是否有统计证据。
run_tsWHAT IT ANSWERS
GRAIN & TIME
KEY FIELDS
run_tsseal_idfamily_idRECOMMENDED JOINS
该表主要按自身业务键使用,目前没有推荐的跨表主键联接。
HOW TO QUERY SAFELY
先用 catalog 确认当前契约;可按 catalog 中列出的 filters 进一步缩小范围;单页最多 5000 行,该表不提供稳定 cursor,结果过大时必须缩小标的、日期或其他筛选。
MACHINE CONTRACT
run_ts, seal_id, family_id—canonical_final_activearchetype→ archetype · stringcohort→ cohort · stringfamily_id→ family_id · stringfdr_pass→ fdr_pass · integerseal_id→ seal_id · stringverdict→ verdict · stringPOINT IN TIME
该资产没有 PIT 契约。普通 rows 端点仍按当前公开读模式返回数据;不要把它表述成历史时点快照。
COPY-READY CALLS
同一数据集可用 HTTP、Python SDK、CLI 或 MCP;Key 始终只通过环境变量/Bearer Header 传递。
curl --fail-with-body -H "Authorization: Bearer $FRIENDS_DATA_API_KEY" "https://api.alphahubs.uk/api/v1/data/datasets/research/daybreak_calibration_verdict/rows?limit=100&start=2026-08-01&end=2026-08-01"from friends_data_api import FriendsDataClient, Settings
async with FriendsDataClient(Settings.from_env()) as client:
page = await client.get_dataset_rows('research', 'daybreak_calibration_verdict', limit=100, start='2026-08-01', end='2026-08-01')
print(page.data)
if page.has_more:
print("next cursor:", page.next_cursor)friends-data get-dataset-rows research daybreak_calibration_verdict --limit 100 --start 2026-08-01 --end 2026-08-01{
"arguments": {
"dataset": "daybreak_calibration_verdict",
"end": "2026-08-01",
"filters": {},
"limit": 100,
"namespace": "research",
"start": "2026-08-01"
},
"tool": "get_dataset_rows"
}FIELD DICTIONARY
descriptionSource 可在 JSON 中区分仓库注释、标准术语表和名称推导说明。
| # | 字段 | ClickHouse 类型 | JSON | 可空 | 角色 | 说明 |
|---|---|---|---|---|---|---|
| 1 | family_id | String | string | no | filter, default_order, warehouse_primary_key | Identifier for family.名称推导 |
| 2 | archetype | LowCardinality(String) | string | no | filter | Source-normalized field: archetype.名称推导 |
| 3 | cohort | LowCardinality(String) | string | no | filter | Source-normalized field: cohort.名称推导 |
| 4 | n_fired | Int32 | integer | no | value | Count of fired.名称推导 |
| 5 | weeks | Int32 | integer | no | value | Source-normalized field: weeks.名称推导 |
| 6 | precision | Float64 | number | no | value | Source-normalized field: precision.名称推导 |
| 7 | base_rate | Float64 | number | no | value | Source-normalized field: base rate.名称推导 |
| 8 | lift | Float64 | number | no | value | Source-normalized field: lift.名称推导 |
| 9 | excess_ci_lo | Float64 | number | no | value | Source-normalized field: excess ci lo.名称推导 |
| 10 | excess_ci_hi | Float64 | number | no | value | Source-normalized field: excess ci hi.名称推导 |
| 11 | p_value | Float64 | number | no | value | Source-normalized field: p value.名称推导 |
| 12 | fdr_pass | UInt8 | integer | no | filter | Source-normalized field: fdr pass.名称推导 |
| 13 | sealed_ci_lo | Float64 | number | no | value | Source-normalized field: sealed ci lo.名称推导 |
| 14 | verdict | LowCardinality(String) | string | no | filter | Source-normalized field: verdict.名称推导 |
| 15 | override_reason | String | string | no | value | Source-normalized field: override reason.名称推导 |
| 16 | git_sha | String | string | no | warehouse_primary_key | Source-normalized field: git sha.名称推导 |
| 17 | seal_id | String | string | no | filter, default_order, warehouse_primary_key | Identifier for seal.名称推导 |
| 18 | run_tsUTC_datetime | DateTime64(3, 'UTC') | string / date-time | no | time, default_order | UTC timestamp for run ts.名称推导 |
| 19 | version | UInt64 | integer | no | provenance | Monotonic row version used for correction-aware reads.标准术语表 |
| 20 | is_deleted | UInt8 | integer | no | value | Soft-delete marker; public serving excludes current tombstones.标准术语表 |
换一个字段名、类型或角色。
结构来自 Friends 只读身份可见的真实 serving relation;权限、过滤、排序、PIT 与分页来自可执行 DatasetSpec。Schema observed at 2026-10-06T18:08:52Z。