flip_api.domain.schemas.private

Classes

`Results`
`OmopData`
`OmopCohortResults`
`TrainingMetrics`	One FL training/evaluation metric point. Mirrors flip-utils' `flip/schemas.py` (the sender).
`TrainingLog`	One row for `POST /model/{id}/logs`: free text XOR a typed round event.
`ProjectApprovalBody`
`ProjectApproval`
`TrustSpecificData`
`AggregatedTrustFieldResult`
`AggregatedFieldResult`
`AggregatedCohortStats`
`FetchedAggregationData`
`TrustTaskResponse`	Response model for a single trust task.
`TaskResultInput`	Input model for submitting a task result.

Module Contents

class flip_api.domain.schemas.private.Results

Bases: pydantic.BaseModel

value: str

count: int

class flip_api.domain.schemas.private.OmopData

Bases: pydantic.BaseModel

name: str

results: list[Results]

class flip_api.domain.schemas.private.OmopCohortResults

Bases: pydantic.BaseModel

query_id: uuid.UUID

trust_id: uuid.UUID

created: str

record_count: int

data: list[OmopData]

error: str | None = None

suppressed: bool = False

ensure_data_is_list(value: Any) → list[Any]

class flip_api.domain.schemas.private.TrainingMetrics

Bases: pydantic.BaseModel

One FL training/evaluation metric point. Mirrors flip-utils’ flip/schemas.py (the sender).

global_round is provenance — always the FL global round the metric was reported in, never overridden. The plot coordinate is the (x_label, x_value) pair, defaulting to the global round on the “Global Rounds” axis — see FLIP#148.

fl_client_name: str

global_round: int

label: str

result: float

x_value: float

x_label: str

classmethod _default_x_value_to_global_round(data: Any) → Any: Backfill a missing/None x_value from global_round (back-compat with old senders).

class flip_api.domain.schemas.private.TrainingLog

Bases: pydantic.BaseModel

One row for POST /model/{id}/logs: free text XOR a typed round event.

Mirrored in flip-utils/flip/schemas.py (the FL-side sender) — keep the two in sync. Two mutually exclusive shapes:

Free text (log set): exception reports and legacy messages. The text is stored and served verbatim.
Typed event (event_type set): a round-progress fact. Display text is composed hub-side at serve time, so the FL layer never bakes in wording.

fl_client_name is the FL kit slot when the row is trust-attributed, or None for hub-attributed rows (e.g. ROUND_STARTED). Old FL images keep sending the original {fl_client_name, log} shape — every new field is optional with a compatible default.

fl_client_name: str | None = None

log: str | None = None

event_type: str | None

global_round: int | None

details: dict[str, Any] | None = None

success: bool = True

_log_xor_event() → TrainingLog

_bound_details() → TrainingLog

class flip_api.domain.schemas.private.ProjectApprovalBody

Bases: pydantic.BaseModel

trusts: list[uuid.UUID]

class flip_api.domain.schemas.private.ProjectApproval

Bases: pydantic.BaseModel

project_id: uuid.UUID

trust_ids: list[uuid.UUID]

class flip_api.domain.schemas.private.TrustSpecificData

Bases: pydantic.BaseModel

record_count: int

data: list[OmopData]

error: str | None = None

suppressed: bool = False

class flip_api.domain.schemas.private.AggregatedTrustFieldResult

Bases: pydantic.BaseModel

data: Any

trust_name: str

trust_id: str

class flip_api.domain.schemas.private.AggregatedFieldResult

Bases: pydantic.BaseModel

name: str

results: list[AggregatedTrustFieldResult]

class flip_api.domain.schemas.private.AggregatedCohortStats

Bases: pydantic.BaseModel

record_count: int

trusts_results: list[AggregatedFieldResult]

trust_record_counts: dict[str, int]

trust_errors: dict[str, str]

trust_suppressed: list[str]

class flip_api.domain.schemas.private.FetchedAggregationData

Bases: pydantic.BaseModel

trust_name: list[str]

trust_id: list[str]

data: list[str]

class flip_api.domain.schemas.private.TrustTaskResponse

Bases: pydantic.BaseModel

Response model for a single trust task.

id: uuid.UUID

task_type: flip_api.domain.schemas.status.TaskType

payload: str

created_at: datetime.datetime

class flip_api.domain.schemas.private.TaskResultInput

Bases: pydantic.BaseModel

Input model for submitting a task result.

success: bool

result: str | None