Aperçu de l’API

`classe` `AggregationType`

Fonctions d’agrégation prises en charge pour les métriques de feedback et de statistiques d’appel.

`classe` `AliasesListReq`

Champs Pydantic :

project_id: <class 'str'>
wb_user_id: str | None

`classe` `AliasesListRes`

Champs Pydantic :

aliases : list[str]

`classe` `AnnotationQueueAddCallsReq`

Requête permettant d’ajouter des appels à une file d’annotation par lot. Étend AnnotationQueueAddCallsBody en ajoutant queue_id pour un usage interne de l’API. Champs Pydantic :

project_id: <classe 'str'>
call_ids: list[str]
display_fields: list[str]
queue_id: <classe 'str'>
wb_user_id: str | None

`classe` `AnnotationQueueAddCallsRes`

Réponse à l’ajout d’appels à une file d’annotation. Champs Pydantic :

added_count: <classe 'int'>
duplicates: <classe 'int'>

`classe` `AnnotationQueueCreateReq`

Requête permettant de créer une nouvelle file d’annotation. Champs Pydantic :

project_id: <classe 'str'>
name: <classe 'str'>
description: <classe 'str'>
scorer_refs: list[str]
wb_user_id: str | None

`classe` `AnnotationQueueCreateRes`

Réponse à la création d’une file d’annotation. Champs Pydantic :

id: <classe 'str'>

`classe` `AnnotationQueueDeleteReq`

Requête permettant de supprimer (suppression réversible) une file d’annotation. Champs Pydantic :

project_id: <classe 'str'>
queue_id: <classe 'str'>
wb_user_id: str | None

`classe` `AnnotationQueueDeleteRes`

Réponse à la suppression d’une file d’annotation. Champs Pydantic :

queue: <classe 'AnnotationQueueSchema'>

`classe` `AnnotationQueueItemSchema`

Schéma des réponses des éléments de la file d’annotation. Champs Pydantic :

id: <class 'str'>
project_id: <class 'str'>
queue_id: <class 'str'>
call_id: <class 'str'>
call_started_at: <class 'datetime.datetime'>
call_ended_at: datetime.datetime | None
call_op_name: <class 'str'>
call_trace_id: <class 'str'>
display_fields: list[str]
added_by: str | None
annotation_state: typing.Literal['unstarted', 'in_progress', 'completed', 'skipped']
annotator_user_id: str | None
created_at: <class 'datetime.datetime'>
created_by: <class 'str'>
updated_at: <class 'datetime.datetime'>
deleted_at: datetime.datetime | None
position_in_queue: int | None

`classe` `AnnotationQueueItemsQueryReq`

Requête permettant d’interroger les éléments d’une file d’annotation. Étend AnnotationQueueItemsQueryBody en ajoutant queue_id pour un usage interne de l’API. Champs Pydantic :

project_id: <classe 'str'>
filter: weave.trace_server.common_interface.AnnotationQueueItemsFilter | None
sort_by: list[weave.trace_server.common_interface.SortBy] | None
limit: int | None
offset: int | None
include_position: <classe 'bool'>
queue_id: <classe 'str'>

`classe` `AnnotationQueueItemsQueryRes`

Réponse à une requête portant sur les éléments de la file d’annotation. Champs Pydantic :

items: list[AnnotationQueueItemSchema]

`classe` `AnnotationQueueReadReq`

Requête permettant de lire une file d’annotation spécifique. Champs Pydantic :

project_id: <classe 'str'>
queue_id: <classe 'str'>

`classe` `AnnotationQueueReadRes`

Réponse à la lecture d’une file d’annotation. Champs Pydantic :

queue: <classe 'AnnotationQueueSchema'>

`classe` `AnnotationQueueSchema`

Schéma des réponses de file d’annotation. Champs Pydantic :

id: <class 'str'>
project_id: <class 'str'>
name: <class 'str'>
description: <class 'str'>
scorer_refs: list[str]
created_at: <class 'datetime.datetime'>
created_by: <class 'str'>
updated_at: <class 'datetime.datetime'>
deleted_at: datetime.datetime | None

`classe` `AnnotationQueueStatsSchema`

Statistiques d’une file d’annotation. Champs Pydantic :

queue_id: <class 'str'>
total_items: <class 'int'>
completed_items: <class 'int'>

`classe` `AnnotationQueueUpdateReq`

Requête permettant de mettre à jour une file d’annotation. Tous les champs sauf project_id et queue_id sont facultatifs : seuls les champs fournis seront mis à jour. champs Pydantic :

project_id: <class 'str'>
queue_id: <class 'str'>
name: str | None
description: str | None
scorer_refs: list[str] | None
wb_user_id: str | None

`classe` `AnnotationQueueUpdateRes`

Réponse à la mise à jour d’une file d’annotation. Champs Pydantic :

queue: <class 'AnnotationQueueSchema'>

`classe` `AnnotationQueuesQueryReq`

Requête permettant d’interroger les files d’annotation d’un projet. Champs Pydantic :

project_id: <class 'str'>
name: str | None
sort_by: list[weave.trace_server.common_interface.SortBy] | None
limit: int | None
offset: int | None

`classe` `AnnotationQueuesQueryRes`

Réponse à une requête sur les files d’annotation. Champs Pydantic :

queues: list[AnnotationQueueSchema]

`classe` `AnnotationQueuesStatsReq`

Requête permettant d’obtenir des statistiques sur plusieurs files d’annotation. Champs Pydantic :

project_id: <class 'str'>
queue_ids: list[str]

`classe` `AnnotationQueuesStatsRes`

Réponse contenant des statistiques sur plusieurs files d’annotation. Champs Pydantic :

stats: list[AnnotationQueueStatsSchema]

`classe` `AnnotatorQueueItemsProgressUpdateReq`

Requête permettant de mettre à jour l’état d’annotation d’un élément de la file d’attente pour l’annotateur actuel. Transitions d’état valides :

(absence) -> ‘in_progress’ : Marquer l’élément comme étant en cours (uniquement lorsqu’aucun enregistrement n’existe)
(absence) -> ‘completed’ ou ‘skipped’ : Terminer ou ignorer directement l’élément
‘in_progress’ ou ‘unstarted’ -> ‘completed’ ou ‘skipped’ : Terminer ou ignorer un élément déjà démarré
same_state -> same_state : Aucune opération idempotente (renvoie l’élément existant sans modification)

Champs Pydantic :

project_id: <class 'str'>
queue_id: <class 'str'>
item_id: <class 'str'>
annotation_state: <class 'str'>
wb_user_id: str | None

`classe` `AnnotatorQueueItemsProgressUpdateRes`

Réponse à la mise à jour de l’état d’annotation. Champs Pydantic :

item: <class 'AnnotationQueueItemSchema'>

`classe` `CallBatchEndMode`

champs Pydantic :

mode: <class 'str'>
req: <class 'CallEndReq'>

`classe` `CallBatchStartMode`

champs Pydantic :

mode: <class 'str'>
req: <class 'CallStartReq'>

`classe` `CallCreateBatchReq`

Champs Pydantic :

batch : list[CallBatchStartMode | CallBatchEndMode]

`classe` `CallCreateBatchRes`

champs Pydantic :

res: list[CallStartRes | CallEndRes]

`classe` `CallEndReq`

champs Pydantic :

end: <class 'EndedCallSchemaForInsert'>

`classe` `CallEndRes`

`classe` `CallEndV2Req`

Requête pour mettre fin à un appel unique via l’API v2. Champs Pydantic :

end: <class 'EndedCallSchemaForInsertWithStartedAt'>

`classe` `CallEndV2Res`

Réponse permettant de terminer un appel unique via l’API v2.

`classe` `CallMetricSpec`

Spécification d’une métrique au niveau des appels à agréger (sans regroupement par modèle). champs Pydantic :

metric: typing.Literal['latency_ms', 'call_count', 'error_count']
aggregations: list[AggregationType]
percentiles: list[float]

`classe` `CallReadReq`

Champs Pydantic :

project_id: <class 'str'>
id: <class 'str'>
include_costs: bool | None
include_storage_size: bool | None
include_total_storage_size: bool | None

`classe` `CallReadRes`

Champs Pydantic :

call : CallSchema | None

`classe` `CallSchema`

Champs Pydantic :

id: <class 'str'>
project_id: <class 'str'>
op_name: <class 'str'>
display_name: str | None
trace_id: <class 'str'>
parent_id: str | None
thread_id: str | None
turn_id: str | None
started_at: <class 'datetime.datetime'>
attributes: dict[str, typing.Any]
inputs: dict[str, typing.Any]
ended_at: datetime.datetime | None
exception: str | None
output: typing.Any | None
summary: SummaryMap | None
wb_user_id: str | None
wb_username: str | None
wb_run_id: str | None
wb_run_step: int | None
wb_run_step_end: int | None
deleted_at: datetime.datetime | None
expire_at: datetime.datetime | None
storage_size_bytes: int | None
total_storage_size_bytes: int | None

`méthode` `serialize_typed_dicts`

serialize_typed_dicts(v: dict[str, Any]) → dict[str, Any]

`classe` `CallStartReq`

Champs Pydantic :

start: <class 'StartedCallSchemaForInsert'>

`classe` `CallStartRes`

champs Pydantic :

id: <class 'str'>
trace_id: <class 'str'>

`classe` `CallStartV2Req`

Requête pour démarrer un appel unique via l’API v2. Champs Pydantic :

start: <class 'StartedCallSchemaForInsert'>

`classe` `CallStartV2Res`

Réponse pour démarrer un appel unique via l’API v2. champs Pydantic :

id: <class 'str'>
trace_id: <class 'str'>

`classe` `CallStatsReq`

Requête de statistiques agrégées sur les appels pour une période donnée. Champs Pydantic :

project_id: <class 'str'>
start: <class 'datetime.datetime'>
end: datetime.datetime | None
granularity: int | None
usage_metrics: list[UsageMetricSpec] | None
call_metrics: list[CallMetricSpec] | None
filter: CallsFilter | None
timezone: <class 'str'>

`méthode` `validate_date_range`

validate_date_range() → CallStatsReq

Veillez à ce que les requêtes de statistiques d’appel soient bornées à une plage de dates sûre.

`classe` `CallStatsRes`

Réponse contenant des statistiques d’appels sous forme de séries temporelles. Champs Pydantic :

start: <class 'datetime.datetime'>
end: <class 'datetime.datetime'>
granularity: <class 'int'>
timezone: <class 'str'>
usage_buckets: list[dict[str, typing.Any]]
call_buckets: list[dict[str, typing.Any]]

`classe` `CallUpdateReq`

Champs Pydantic :

project_id: <class 'str'>
call_id: <class 'str'>
display_name: str | None
wb_user_id: str | None

`classe` `CallUpdateRes`

`classe` `CallsDeleteReq`

Champs Pydantic :

project_id: <class 'str'>
call_ids: list[str]
wb_user_id: str | None

`classe` `CallsDeleteRes`

champs Pydantic :

num_deleted: <class 'int'>

`classe` `CallsFilter`

Champs Pydantic :

op_names: list[str] | None
input_refs: list[str] | None
output_refs: list[str] | None
parent_ids: list[str] | None
trace_ids: list[str] | None
call_ids: list[str] | None
thread_ids: list[str] | None
turn_ids: list[str] | None
trace_roots_only: bool | None
wb_user_ids: list[str] | None
wb_run_ids: list[str] | None

`classe` `CallsQueryReq`

Champs Pydantic :

project_id: <class 'str'>
filter: CallsFilter | None
limit: int | None
offset: int | None
sort_by: list[weave.trace_server.common_interface.SortBy] | None
query: weave.trace_server.interface.query.Query | None
include_costs: bool | None
include_feedback: bool | None
include_storage_size: bool | None
include_total_storage_size: bool | None
include_usernames: bool | None
columns: list[str] | None
expand_columns: list[str] | None
return_expanded_column_values: bool | None

`classe` `CallsQueryRes`

Champs Pydantic :

calls : list[CallSchema]

`classe` `CallsQueryStatsReq`

Champs Pydantic :

project_id: <class 'str'>
filter: CallsFilter | None
query: weave.trace_server.interface.query.Query | None
limit: int | None
include_total_storage_size: bool | None
expand_columns: list[str] | None

`classe` `CallsQueryStatsRes`

champs Pydantic :

count: <class 'int'>
has_more: <class 'bool'>
total_storage_size_bytes: int | None

`classe` `CallsScoreReq`

Requête permettant de mettre en file d’attente des jobs de scoring pour une liste d’appels. Le scoring est effectué de manière asynchrone par le call_scoring_worker, qui consomme les messages de Kafka et applique chaque scorer_ref à chaque call_id. Champs Pydantic :

project_id: <class 'str'>
call_ids: list[str]
scorer_refs: list[str]
wb_user_id: str | None

`classe` `CallsScoreRes`

Réponse vide pour calls_score. Définie comme un modèle (plutôt qu’en renvoyant None) afin de respecter la convention utilisée dans toute cette interface et de permettre l’ajout ultérieur de champs sans changement incompatible.

`classe` `CallsUpsertCompleteReq`

Requête de mise à jour ou insertion d’un lot d’appels terminés. Champs Pydantic :

batch: list[CompletedCallSchemaForInsert]

`classe` `CallsUpsertCompleteRes`

Réponse pour la mise à jour ou l’insertion d’un lot d’appels terminés.

`classe` `CallsUsageReq`

Requête permettant de calculer l’utilisation agrégée de plusieurs appels racine. Ce point de terminaison renvoie les métriques d’utilisation pour chaque appel racine demandé. Les métriques de chaque racine incluent la somme de sa propre utilisation et de celle de tous ses descendants. Remarque : tous les appels correspondants sont chargés en mémoire pour l’agrégation. Pour les jeux de résultats très volumineux (>10k appels), envisagez de traiter les ID d’appels racine par lots ou d’utiliser des filtres plus restrictifs au niveau de l’application. Champs Pydantic :

project_id: <class 'str'>
call_ids: list[str]
include_costs: <class 'bool'>
limit: <class 'int'>

`classe` `CallsUsageRes`

Réponse contenant des métriques d’utilisation agrégées par appel racine. Champs Pydantic :

call_usage: dict[str, dict[str, LLMAggregatedUsage]]
unfinished_call_ids: list[str]

`classe` `CompletedCallSchemaForInsert`

Schéma permettant d’insérer directement un appel terminé. Représente un appel déjà terminé au moment de l’insertion, avec les informations de début et de fin fournies ensemble. Utilisé par le point de terminaison calls_complete. Champs Pydantic :

project_id: <class 'str'>
id: <class 'str'>
trace_id: <class 'str'>
op_name: <class 'str'>
started_at: <class 'datetime.datetime'>
ended_at: <class 'datetime.datetime'>
display_name: str | None
parent_id: str | None
thread_id: str | None
turn_id: str | None
attributes: dict[str, typing.Any]
inputs: dict[str, typing.Any]
output: typing.Any | None
summary: <class 'SummaryInsertMap'>
otel_dump: dict[str, typing.Any] | None
exception: str | None
wb_user_id: str | None
wb_run_id: str | None
wb_run_step: int | None
wb_run_step_end: int | None

`méthode` `serialize_typed_dicts`

serialize_typed_dicts(v: dict[str, Any]) → dict[str, Any]

`classe` `CompletionsCreateReq`

Champs Pydantic :

project_id: <class 'str'>
inputs: <class 'CompletionsCreateRequestInputs'>
wb_user_id: str | None
track_llm_call: bool | None
trace_id: str | None
parent_id: str | None
conversation_id: str | None
conversation_name: str | None
source: str | None

`classe` `CompletionsCreateRequestInputs`

Champs Pydantic :

model: <class 'str'>
messages: <class 'list'>
timeout: float | str | None
temperature: float | None
top_p: float | None
n: int | None
stop: str | list | None
max_completion_tokens: int | None
max_tokens: int | None
modalities: list | None
presence_penalty: float | None
frequency_penalty: float | None
stream: bool | None
logit_bias: dict | None
user: str | None
response_format: dict | type[pydantic.main.BaseModel] | None
seed: int | None
tools: list | None
tool_choice: str | dict | None
logprobs: bool | None
top_logprobs: int | None
parallel_tool_calls: bool | None
extra_headers: dict | None
functions: list | None
function_call: str | None
api_version: str | None
prompt: str | None
template_vars: dict[str, typing.Any] | None
vertex_credentials: str | None

`classe` `CompletionsCreateRes`

Champs Pydantic :

response: dict[str, typing.Any]
weave_call_id: str | None
span_id: str | None
trace_id: str | None
conversation_id: str | None

`classe` `CostCreateInput`

Champs Pydantic :

prompt_token_cost: <class 'float'>
completion_token_cost: <class 'float'>
cache_read_input_token_cost: <class 'float'>
cache_creation_input_token_cost: <class 'float'>
prompt_token_cost_unit: str | None
completion_token_cost_unit: str | None
effective_date: datetime.datetime | None
provider_id: str | None

`classe` `CostCreateReq`

Champs Pydantic :

project_id: <class 'str'>
costs: dict[str, CostCreateInput]
wb_user_id: str | None

`classe` `CostCreateRes`

Champs Pydantic :

ids: list[tuple[str, str]]

`classe` `CostPurgeReq`

Champs Pydantic :

project_id: <class 'str'>
query: <class 'weave.trace_server.interface.query.Query'>

`classe` `CostPurgeRes`

`classe` `CostQueryOutput`

Champs Pydantic :

id: str | None
llm_id: str | None
prompt_token_cost: float | None
completion_token_cost: float | None
prompt_token_cost_unit: str | None
completion_token_cost_unit: str | None
effective_date: datetime.datetime | None
provider_id: str | None

`classe` `CostQueryReq`

Champs Pydantic :

project_id: <class 'str'>
fields: list[str] | None
query: weave.trace_server.interface.query.Query | None
sort_by: list[weave.trace_server.common_interface.SortBy] | None
limit: int | None
offset: int | None

`classe` `CostQueryRes`

Champs Pydantic :

results: list[CostQueryOutput]

`classe` `DatasetCreateBody`

Champs Pydantic :

name: str | None
description: str | None
rows: list[dict[str, typing.Any]]

`classe` `DatasetCreateReq`

Champs Pydantic :

name: str | None
description: str | None
rows: list[dict[str, typing.Any]]
project_id: <class 'str'>
wb_user_id: str | None

`classe` `DatasetCreateRes`

Champs Pydantic :

digest: <class 'str'>
object_id: <class 'str'>
version_index: <class 'int'>

`classe` `DatasetDeleteReq`

champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digests: list[str] | None
wb_user_id: str | None

`classe` `DatasetDeleteRes`

Champs Pydantic :

num_deleted: <class 'int'>

`classe` `DatasetListReq`

Champs Pydantic :

project_id: <class 'str'>
limit: int | None
offset: int | None
wb_user_id: str | None

`classe` `DatasetReadReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digest: <class 'str'>
wb_user_id: str | None

`classe` `DatasetReadRes`

Champs Pydantic :

object_id: <class 'str'>
digest: <class 'str'>
version_index: <class 'int'>
created_at: <class 'datetime.datetime'>
name: <class 'str'>
description: str | None
rows: <class 'str'>

`classe` `DatasetSourceLinkPayload`

Une seule ligne de dataset et les sources à lui associer. Champs Pydantic :

row_digest: <class 'str'>
sources: list[SourceRef]
link_metadata: dict[str, typing.Any] | None

`classe` `DatasetSourceLinkSchema`

Schéma d’une seule ligne de lien vers une source de dataset. Champs Pydantic :

id: <class 'str'>
row_digest: <class 'str'>
source_kind: <enum 'SourceKind'>
source_id: <class 'str'>
source_trace_id: <class 'str'>
source_started_at: <class 'datetime.datetime'>
source_display_name: <class 'str'>
link_metadata: dict[str, typing.Any] | None
added_by: str | None
created_at: <class 'datetime.datetime'>
updated_at: <class 'datetime.datetime'>
deleted_at: datetime.datetime | None

`classe` `DatasetSourcesLinkDeleteReq`

Requête permettant de supprimer de manière réversible les liens de source du dataset par ID. Champs Pydantic :

project_id: <class 'str'>
link_ids: list[str]
wb_user_id: str | None

`classe` `DatasetSourcesLinkDeleteRes`

Réponse à la suppression des liens vers les sources du dataset. Champs Pydantic :

entries: list[DatasetSourcesLinkDeleteResEntry]

`classe` `DatasetSourcesLinkDeleteResEntry`

Résultat de la suppression d’un seul lien. Champs Pydantic :

link_id: <class 'str'>
deleted: <class 'bool'>

`classe` `DatasetSourcesLinkReq`

Requête permettant de lier les lignes du dataset à leurs sources d’origine. Champs Pydantic :

project_id: <class 'str'>
dataset_object_id: <class 'str'>
dataset_digest: <class 'str'>
links: list[DatasetSourceLinkPayload]
include_created_status: <class 'bool'>
wb_user_id: str | None

`classe` `DatasetSourcesLinkRes`

Réponse à l’association des lignes du dataset aux sources. Une entrée pour chaque tuple (row_digest, source) aplati, dans l’ordre des entrées. champs Pydantic :

entries: list[DatasetSourcesLinkResEntry]

`classe` `DatasetSourcesLinkResEntry`

Résultat pour un seul lien (row_digest, source) mis à plat. Champs Pydantic :

link_id: <class 'str'>
created: bool | None

`classe` `DatasetSourcesQueryReq`

Requête dans le sens direct : dataset -> sources. Champs Pydantic :

project_id: <class 'str'>
dataset_object_id: <class 'str'>
row_digests: list[str] | None
source_kinds: list[SourceKind] | None
include_deleted: <class 'bool'>
limit: int | None
offset: int | None
wb_user_id: str | None

`classe` `DatasetSourcesQueryRes`

Réponse à une requête directe dataset -> sources. Champs Pydantic :

links: list[DatasetSourceLinkSchema]

`classe` `EndedCallSchemaForInsert`

champs Pydantic :

project_id: <class 'str'>
id: <class 'str'>
trace_id: str | None
is_eval: bool | None
ended_at: <class 'datetime.datetime'>
started_at: datetime.datetime | None
exception: str | None
output: typing.Any | None
summary: <class 'SummaryInsertMap'>
wb_run_step_end: int | None

`méthode` `serialize_typed_dicts`

serialize_typed_dicts(v: dict[str, Any]) → dict[str, Any]

`classe` `EndedCallSchemaForInsertWithStartedAt`

Alias obsolète. started_at se trouve désormais sur le parent EndedCallSchemaForInsert ; privilégiez celui-ci. Conservé pour que les SDK externes épinglés sur le nom WithStartedAt continuent à l’importer. À supprimer une fois que tous les appelants internes à l’arborescence auront migré. champs Pydantic :

project_id: <class 'str'>
id: <class 'str'>
trace_id: str | None
is_eval: bool | None
ended_at: <class 'datetime.datetime'>
started_at: datetime.datetime | None
exception: str | None
output: typing.Any | None
summary: <class 'SummaryInsertMap'>
wb_run_step_end: int | None

`méthode` `serialize_typed_dicts`

serialize_typed_dicts(v: dict[str, Any]) → dict[str, Any]

`classe` `EvalResultsEvaluationSummary`

champs Pydantic :

evaluation_call_id: <class 'str'>
trial_count: <class 'int'>
scorer_stats: list[EvalResultsScorerStats]
predict_total_tokens: int | None
predict_total_cost: float | None
evaluation_ref: str | None
model_ref: str | None
display_name: str | None
trace_id: str | None
started_at: str | None

`classe` `EvalResultsFilter`

Un filtre limité à une évaluation facultative. Champs Pydantic :

evaluation_call_id: str | None
query: <class 'weave.trace_server.interface.query.Query'>

`classe` `EvalResultsQueryBody`

Champs Pydantic :

evaluation_call_ids: list[str] | None
evaluation_run_ids: list[str] | None
require_intersection: <class 'bool'>
include_raw_data_rows: <class 'bool'>
resolve_row_refs: <class 'bool'>
include_rows: <class 'bool'>
include_summary: <class 'bool'>
summary_require_intersection: bool | None
include_predict_and_score_children: <class 'bool'>
include_costs: <class 'bool'>
sort_by: list[EvalResultsSortBy] | None
filters: list[EvalResultsFilter] | None
filter_logic_operator: typing.Literal['and', 'or']
limit: int | None
offset: <class 'int'>

`méthode` `validate_identifiers`

validate_identifiers() → EvalResultsQueryBody

Vérifiez qu’au moins un identifiant d’évaluation est fourni.

`classe` `EvalResultsQueryReq`

Champs Pydantic :

evaluation_call_ids: list[str] | None
evaluation_run_ids: list[str] | None
require_intersection: <class 'bool'>
include_raw_data_rows: <class 'bool'>
resolve_row_refs: <class 'bool'>
include_rows: <class 'bool'>
include_summary: <class 'bool'>
summary_require_intersection: bool | None
include_predict_and_score_children: <class 'bool'>
include_costs: <class 'bool'>
sort_by: list[EvalResultsSortBy] | None
filters: list[EvalResultsFilter] | None
filter_logic_operator: typing.Literal['and', 'or']
limit: int | None
offset: <class 'int'>
project_id: <class 'str'>

`méthode` `validate_identifiers`

validate_identifiers() → EvalResultsQueryBody

Vérifiez qu’au moins un identifiant d’évaluation est fourni.

`classe` `EvalResultsQueryRes`

Champs Pydantic :

rows: list[EvalResultsRow]
total_rows: <class 'int'>
summary: ForwardRef('EvalResultsSummaryRes | None')
warnings: list[str]

`classe` `EvalResultsRow`

Champs Pydantic :

row_digest: <class 'str'>
raw_data_row: typing.Any | None
evaluations: list[EvalResultsRowEvaluation]

`classe` `EvalResultsRowEvaluation`

Champs Pydantic :

evaluation_call_id: <class 'str'>
trials: list[EvalResultsTrial]

`classe` `EvalResultsScorerStats`

Statistiques pour une seule dimension de score mise à plat (scorer_key ou scorer_key.path.to.leaf). Champs Pydantic :

scorer_key: <class 'str'>
path: str | None
value_type: typing.Optional[typing.Literal['binary', 'continuous', 'text']]
trial_count: <class 'int'>
numeric_count: <class 'int'>
numeric_mean: float | None
pass_true_count: <class 'int'>
pass_known_count: <class 'int'>
pass_rate: float | None
pass_signal_coverage: float | None

`classe` `EvalResultsSortBy`

Spécification de tri pour les résultats d’évaluation, qui étend SortBy Champs Pydantic :

field: <class 'str'>
direction: typing.Literal['asc', 'desc']
evaluation_call_id: str | None
mode: typing.Literal['value', 'difference']

`classe` `EvalResultsSummaryRes`

Champs Pydantic :

row_count: <class 'int'>
evaluations: list[EvalResultsEvaluationSummary]

`classe` `EvalResultsTrial`

Champs Pydantic :

predict_and_score_call_id: <class 'str'>
predict_call_id: str | None
model_output: typing.Any | None
scores: dict[str, typing.Any]
model_latency_seconds: float | None
total_tokens: int | None
total_cost: float | None
scorer_call_ids: dict[str, str]
genai_span_ref: list[GenAISpanRef] | None

`classe` `EvaluateModelArgs`

Arguments d’une tâche evaluate-model complète (charge le modèle + exécute les prédictions + calcule les scores). Déplacé depuis workers/evaluate_model_worker/evaluate_model_worker.py afin que les deux types de tâche (EvaluateModelArgs et RescoringArgs) soient regroupés dans le même module pour l’union discriminée EvalWorkerJob. champs Pydantic :

job_type: typing.Literal['evaluate_model']
project_id: <class 'str'>
evaluation_ref: <class 'str'>
model_ref: <class 'str'>
wb_user_id: <class 'str'>
evaluation_call_id: <class 'str'>

`classe` `EvaluateModelReq`

Champs Pydantic :

project_id: <class 'str'>
evaluation_ref: <class 'str'>
model_ref: <class 'str'>
wb_user_id: str | None

`classe` `EvaluateModelRes`

Champs Pydantic :

call_id: <class 'str'>

`classe` `EvaluationCreateBody`

Champs Pydantic :

name: <class 'str'>
description: str | None
dataset: <class 'str'>
scorers: list[str] | None
trials: <class 'int'>
evaluation_name: str | None
eval_attributes: dict[str, typing.Any] | None

`classe` `EvaluationCreateReq`

champs Pydantic :

name: <class 'str'>
description: str | None
dataset: <class 'str'>
scorers: list[str] | None
trials: <class 'int'>
evaluation_name: str | None
eval_attributes: dict[str, typing.Any] | None
project_id: <class 'str'>
wb_user_id: str | None

`classe` `EvaluationCreateRes`

Champs Pydantic :

digest: <class 'str'>
object_id: <class 'str'>
version_index: <class 'int'>
evaluation_ref: <class 'str'>

`classe` `EvaluationDeleteReq`

champs Pydantic :

project_id : <class 'str'>
object_id : <class 'str'>
digests : list[str] | None
wb_user_id : str | None

`classe` `EvaluationDeleteRes`

Champs Pydantic :

num_deleted: <class 'int'>

`classe` `EvaluationListReq`

Champs Pydantic :

project_id: <class 'str'>
limit: int | None
offset: int | None
wb_user_id: str | None

`classe` `EvaluationReadReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digest: <class 'str'>
wb_user_id: str | None

`classe` `EvaluationReadRes`

Champs Pydantic :

object_id: <class 'str'>
digest: <class 'str'>
version_index: <class 'int'>
created_at: <class 'datetime.datetime'>
name: <class 'str'>
description: str | None
dataset: <class 'str'>
scorers: list[str]
trials: <class 'int'>
evaluation_name: str | None
evaluate_op: str | None
predict_and_score_op: str | None
summarize_op: str | None

`classe` `EvaluationRunCreateBody`

Champs Pydantic :

evaluation: <class 'str'>
model: <class 'str'>
source_evaluation_run_id: str | None

`classe` `EvaluationRunCreateReq`

champs Pydantic :

evaluation: <class 'str'>
model: <class 'str'>
source_evaluation_run_id: str | None
project_id: <class 'str'>
wb_user_id: str | None

`classe` `EvaluationRunCreateRes`

champs Pydantic :

evaluation_run_id: <class 'str'>

`classe` `EvaluationRunDeleteReq`

Champs Pydantic :

project_id: <class 'str'>
evaluation_run_ids: list[str]
wb_user_id: str | None

`classe` `EvaluationRunDeleteRes`

Champs Pydantic :

num_deleted: <class 'int'>

`classe` `EvaluationRunFilter`

Champs Pydantic :

evaluations: list[str] | None
models: list[str] | None
evaluation_run_ids: list[str] | None

`classe` `EvaluationRunFinishBody`

Corps de requête pour terminer un run d’évaluation via l’API REST. Ce modèle exclut project_id et evaluation_run_id, car ils proviennent du chemin de l’URL dans les points de terminaison REST. Champs Pydantic :

summary: dict[str, typing.Any] | None

`classe` `EvaluationRunFinishReq`

Champs Pydantic :

summary: dict[str, typing.Any] | None
project_id: <class 'str'>
evaluation_run_id: <class 'str'>
wb_user_id: str | None

`classe` `EvaluationRunFinishRes`

champs Pydantic :

success : <class 'bool'>

`classe` `EvaluationRunListReq`

Champs Pydantic :

project_id: <class 'str'>
filter: EvaluationRunFilter | None
limit: int | None
offset: int | None

`classe` `EvaluationRunReadReq`

Champs Pydantic :

project_id: <class 'str'>
evaluation_run_id: <class 'str'>

`classe` `EvaluationRunReadRes`

Champs Pydantic :

evaluation_run_id: <class 'str'>
evaluation: <class 'str'>
model: <class 'str'>
status: str | None
started_at: datetime.datetime | None
finished_at: datetime.datetime | None
summary: dict[str, typing.Any] | None
source_evaluation_run_id: str | None

`classe` `EvaluationStatusComplete`

Champs Pydantic :

code: typing.Literal['complete']
output: dict[str, typing.Any]

`classe` `EvaluationStatusFailed`

champs Pydantic :

code: typing.Literal['failed']
error: str | None

`classe` `EvaluationStatusNotFound`

champs Pydantic :

code: typing.Literal['not_found']

`classe` `EvaluationStatusReq`

Champs Pydantic :

project_id: <class 'str'>
call_id: <class 'str'>

`classe` `EvaluationStatusRes`

Champs Pydantic :

status : EvaluationStatusNotFound | EvaluationStatusRunning | EvaluationStatusFailed | EvaluationStatusComplete

`classe` `EvaluationStatusRunning`

champs Pydantic :

code: typing.Literal['running']
completed_rows: <class 'int'>
total_rows: <class 'int'>

`classe` `ExportTracePartialSuccess`

champs Pydantic :

rejected_spans: <class 'int'>
error_message: <class 'str'>

`classe` `ExtraKeysTypedDict`

`classe` `Feedback`

Champs Pydantic :

id: <class 'str'>
project_id: <class 'str'>
weave_ref: <class 'str'>
creator: str | None
feedback_type: <class 'str'>
payload: dict[str, typing.Any]
annotation_ref: str | None
runnable_ref: str | None
call_ref: str | None
trigger_ref: str | None
queue_id: str | None
scorer_tags: list[str]
scorer_tag_reasons: dict[str, str]
scorer_tag_confidences: dict[str, float]
scorer_ratings: dict[str, float]
scorer_rating_reasons: dict[str, str]
scorer_rating_confidences: dict[str, float]
span_agent_name: <class 'str'>
span_agent_version: <class 'str'>
span_status_code: <class 'str'>
wb_user_id: str | None
created_at: <class 'datetime.datetime'>

`classe` `FeedbackAggregateBucket`

Une ligne (tranche temporelle, groupe) de feedback agrégé provenant du scorer. Champs Pydantic :

time_bucket_start_ms: int | None
group: dict[str, str]
total_count: <class 'int'>
scored_count: <class 'int'>
tag_counts: dict[str, int]
rating_counts: dict[str, int]
rating_sums: dict[str, float]

`classe` `FeedbackAggregateReq`

Requête de scores agrégés par tranche temporelle et par dimension. Champs Pydantic :

project_id: <class 'str'>
after_ms: <class 'int'>
before_ms: <class 'int'>
time_bucket_seconds: int | None
feedback_types: list[str]
tags: list[str]
rating_min: float | None
rating_max: float | None
monitor_ids: list[str]
scorer_ids: list[str]
span_agent_names: list[str]
span_types: list[typing.Literal['agent_turn', 'agent_conversation']]
group_by: list[typing.Literal['scorer_id', 'span_agent_name', 'span_agent_version', 'span_status_code']]

`classe` `FeedbackAggregateRes`

Série temporelle clairsemée du feedback agrégé du scorer (buckets vides omis). Champs Pydantic :

time_bucket_seconds: int | None
after_ms: <class 'int'>
before_ms: <class 'int'>
buckets: list[FeedbackAggregateBucket]

`classe` `FeedbackCreateBatchReq`

champs Pydantic :

batch: list[FeedbackCreateReq]

`classe` `FeedbackCreateBatchRes`

Champs Pydantic :

res: list[FeedbackCreateRes]

`classe` `FeedbackCreateReq`

Champs Pydantic :

id: str | None
project_id: <class 'str'>
weave_ref: <class 'str'>
creator: str | None
feedback_type: <class 'str'>
payload: dict[str, typing.Any]
annotation_ref: str | None
runnable_ref: str | None
call_ref: str | None
trigger_ref: str | None
queue_id: str | None
scorer_tags: list[str]
scorer_tag_reasons: dict[str, str]
scorer_tag_confidences: dict[str, float]
scorer_ratings: dict[str, float]
scorer_rating_reasons: dict[str, str]
scorer_rating_confidences: dict[str, float]
span_agent_name: <class 'str'>
span_agent_version: <class 'str'>
span_status_code: <class 'str'>
wb_user_id: str | None

`classe` `FeedbackCreateRes`

Champs Pydantic :

id: <class 'str'>
created_at: <class 'datetime.datetime'>
wb_user_id: <class 'str'>
payload: dict[str, typing.Any]

`classe` `FeedbackDict`

`classe` `FeedbackMetricSpec`

Spécification d’une métrique de la charge utile de feedback à agréger. Champs Pydantic :

json_path: <class 'str'>
value_type: typing.Literal['numeric', 'boolean', 'categorical']
aggregations: list[AggregationType]
percentiles: list[float]

`classe` `FeedbackPayloadPath`

Chemin découvert dans la charge utile de feedback avec un type inféré. Champs Pydantic :

json_path: <class 'str'>
value_type: typing.Literal['numeric', 'boolean', 'categorical']

`classe` `FeedbackPayloadSchemaReq`

Requête pour la découverte du schéma de charge utile de feedback. Champs Pydantic :

project_id: <class 'str'>
start: <class 'datetime.datetime'>
end: datetime.datetime | None
feedback_type: str | None
trigger_ref: str | None
sample_limit: <class 'int'>

`méthode` `validate_date_range`

validate_date_range() → _FeedbackFilterBase

Assurez-vous que les requêtes de feedback restent limitées à une plage de dates sûre.

`classe` `FeedbackPayloadSchemaRes`

Réponse contenant les chemins et les types détectés de la charge utile de feedback. Champs Pydantic :

paths: list[FeedbackPayloadPath]

`classe` `FeedbackPurgeReq`

champs Pydantic :

project_id: <class 'str'>
query: <class 'weave.trace_server.interface.query.Query'>

`classe` `FeedbackPurgeRes`

`classe` `FeedbackQueryReq`

Champs Pydantic :

project_id: <class 'str'>
fields: list[str] | None
query: weave.trace_server.interface.query.Query | None
sort_by: list[weave.trace_server.common_interface.SortBy] | None
limit: int | None
offset: int | None

`classe` `FeedbackQueryRes`

Champs Pydantic :

result: list[dict[str, typing.Any]]

`classe` `FeedbackReplaceReq`

Champs Pydantic :

id: str | None
project_id: <class 'str'>
weave_ref: <class 'str'>
creator: str | None
feedback_type: <<class 'str'>
payload: dict[str, typing.Any]
annotation_ref: str | None
runnable_ref: str | None
call_ref: str | None
trigger_ref: str | None
queue_id: str | None
scorer_tags: list[str]
scorer_tag_reasons: dict[str, str]
scorer_tag_confidences: dict[str, float]
scorer_ratings: dict[str, float]
scorer_rating_reasons: dict[str, str]
scorer_rating_confidences: dict[str, float]
span_agent_name: <class 'str'>
span_agent_version: <class 'str'>
span_status_code: <class 'str'>
wb_user_id: str | None
feedback_id: <class 'str'>

`classe` `FeedbackReplaceRes`

Champs Pydantic :

id: <class 'str'>
created_at: <class 'datetime.datetime'>
wb_user_id: <class 'str'>
payload: dict[str, typing.Any]

`classe` `FeedbackStatsReq`

Requête pour des statistiques agrégées de feedback par tranche temporelle. Champs Pydantic :

project_id: <class 'str'>
start: <class 'datetime.datetime'>
end: datetime.datetime | None
feedback_type: str | None
trigger_ref: str | None
granularity: int | None
timezone: <class 'str'>
metrics: list[FeedbackMetricSpec]

`méthode` `validate_date_range`

validate_date_range() → _FeedbackFilterBase

Assurez-vous que les requêtes de feedback restent limitées à une plage de dates sûre.

`classe` `FeedbackStatsRes`

Réponse contenant des statistiques de feedback sous forme de série temporelle. Champs Pydantic :

start: <class 'datetime.datetime'>
end: <class 'datetime.datetime'>
granularity: <class 'int'>
timezone: <class 'str'>
buckets: list[dict[str, typing.Any]]
window_stats: dict[str, dict[str, float | None]] | None

`classe` `FileContentReadReq`

Champs Pydantic :

project_id: <class 'str'>
digest: <class 'str'>

`classe` `FileContentReadRes`

Champs Pydantic :

content: <class 'bytes'>

`classe` `FileCreateReq`

Champs Pydantic :

project_id: <class 'str'>
name: <class 'str'>
content: <class 'bytes'>
expected_digest: str | None

`classe` `FileCreateRes`

Champs Pydantic :

digest: <class 'str'>

`classe` `FilesStatsReq`

Champs Pydantic :

project_id: <class 'str'>

`classe` `FilesStatsRes`

Champs Pydantic :

total_size_bytes: <class 'int'>

`classe` `FullTraceServerInterface`

Interface complète du serveur de traces prenant en charge les API V1 et Object. Ce protocole représente une implémentation de serveur de traces qui prend en charge l’ensemble des API, à savoir les endpoints V1 hérités et les endpoints Object modernes. Utilisez ce type pour les implémentations qui doivent prendre en charge les deux versions de l’API.

`méthode` `agent_agents_query`

agent_agents_query(req: AgentsQueryReq) → AgentsQueryRes

`méthode` `agent_conversation_chat`

agent_conversation_chat(
    req: AgentConversationChatReq
) → AgentConversationChatRes

`méthode` `agent_conversation_spans`

agent_conversation_spans(
    req: AgentConversationSpansReq
) → AgentConversationSpansRes

`méthode` `agent_custom_attrs_schema`

agent_custom_attrs_schema(
    req: AgentCustomAttrsSchemaReq
) → AgentCustomAttrsSchemaRes

`méthode` `agent_search`

agent_search(req: AgentSearchReq) → AgentSearchRes

`méthode` `agent_spans_query`

agent_spans_query(req: AgentSpansQueryReq) → AgentSpansQueryRes

`méthode` `agent_spans_stats`

agent_spans_stats(req: AgentSpanStatsReq) → AgentSpanStatsRes

`méthode` `agent_traces_chat`

agent_traces_chat(req: AgentTraceChatReq) → AgentTraceChatRes

`méthode` `agent_versions_query`

agent_versions_query(req: AgentVersionsQueryReq) → AgentVersionsQueryRes

`méthode` `aliases_list`

aliases_list(req: AliasesListReq) → AliasesListRes

`méthode` `annotation_queue_add_calls`

annotation_queue_add_calls(
    req: AnnotationQueueAddCallsReq
) → AnnotationQueueAddCallsRes

`méthode` `annotation_queue_create`

annotation_queue_create(
    req: AnnotationQueueCreateReq
) → AnnotationQueueCreateRes

`méthode` `annotation_queue_delete`

annotation_queue_delete(
    req: AnnotationQueueDeleteReq
) → AnnotationQueueDeleteRes

`méthode` `annotation_queue_items_query`

annotation_queue_items_query(
    req: AnnotationQueueItemsQueryReq
) → AnnotationQueueItemsQueryRes

`méthode` `annotation_queue_read`

annotation_queue_read(req: AnnotationQueueReadReq) → AnnotationQueueReadRes

`méthode` `annotation_queue_update`

annotation_queue_update(
    req: AnnotationQueueUpdateReq
) → AnnotationQueueUpdateRes

`méthode` `annotation_queues_query_stream`

annotation_queues_query_stream(
    req: AnnotationQueuesQueryReq
) → Iterator[AnnotationQueueSchema]

`méthode` `annotation_queues_stats`

annotation_queues_stats(
    req: AnnotationQueuesStatsReq
) → AnnotationQueuesStatsRes

`méthode` `annotator_queue_items_progress_update`

annotator_queue_items_progress_update(
    req: AnnotatorQueueItemsProgressUpdateReq
) → AnnotatorQueueItemsProgressUpdateRes

`méthode` `call_end`

call_end(req: CallEndReq) → CallEndRes

`méthode` `call_end_v2`

call_end_v2(req: CallEndV2Req) → CallEndV2Res

`méthode` `call_read`

call_read(req: CallReadReq) → CallReadRes

`méthode` `call_start`

call_start(req: CallStartReq) → CallStartRes

`méthode` `call_start_batch`

call_start_batch(req: CallCreateBatchReq) → CallCreateBatchRes

`méthode` `call_start_v2`

call_start_v2(req: CallStartV2Req) → CallStartV2Res

`méthode` `call_stats`

call_stats(req: 'CallStatsReq') → CallStatsRes

`méthode` `call_update`

call_update(req: CallUpdateReq) → CallUpdateRes

`méthode` `calls_complete`

calls_complete(req: CallsUpsertCompleteReq) → CallsUpsertCompleteRes

`méthode` `calls_delete`

calls_delete(req: CallsDeleteReq) → CallsDeleteRes

`méthode` `calls_query`

calls_query(req: CallsQueryReq) → CallsQueryRes

`méthode` `calls_query_stats`

calls_query_stats(req: CallsQueryStatsReq) → CallsQueryStatsRes

`méthode` `calls_query_stream`

calls_query_stream(req: CallsQueryReq) → Iterator[CallSchema]

`méthode` `calls_score`

calls_score(req: CallsScoreReq) → CallsScoreRes

`méthode` `calls_usage`

calls_usage(req: 'CallsUsageReq') → CallsUsageRes

`méthode` `completions_create`

completions_create(req: CompletionsCreateReq) → CompletionsCreateRes

`méthode` `completions_create_stream`

completions_create_stream(req: CompletionsCreateReq) → Iterator[dict[str, Any]]

`méthode` `cost_create`

cost_create(req: CostCreateReq) → CostCreateRes

`méthode` `cost_purge`

cost_purge(req: CostPurgeReq) → CostPurgeRes

`méthode` `cost_query`

cost_query(req: CostQueryReq) → CostQueryRes

`méthode` `dataset_create`

dataset_create(req: DatasetCreateReq) → DatasetCreateRes

`méthode` `dataset_delete`

dataset_delete(req: DatasetDeleteReq) → DatasetDeleteRes

`méthode` `dataset_list`

dataset_list(req: DatasetListReq) → Iterator[DatasetReadRes]

`méthode` `dataset_read`

dataset_read(req: DatasetReadReq) → DatasetReadRes

`méthode` `dataset_sources_link`

dataset_sources_link(req: DatasetSourcesLinkReq) → DatasetSourcesLinkRes

`méthode` `dataset_sources_link_delete`

dataset_sources_link_delete(
    req: DatasetSourcesLinkDeleteReq
) → DatasetSourcesLinkDeleteRes

`méthode` `dataset_sources_query`

dataset_sources_query(req: DatasetSourcesQueryReq) → DatasetSourcesQueryRes

`méthode` `eval_results_query`

eval_results_query(req: EvalResultsQueryReq) → EvalResultsQueryRes

`méthode` `evaluate_model`

evaluate_model(req: EvaluateModelReq) → EvaluateModelRes

`méthode` `evaluation_create`

evaluation_create(req: EvaluationCreateReq) → EvaluationCreateRes

`méthode` `evaluation_delete`

evaluation_delete(req: EvaluationDeleteReq) → EvaluationDeleteRes

`méthode` `evaluation_list`

evaluation_list(req: EvaluationListReq) → Iterator[EvaluationReadRes]

`méthode` `evaluation_read`

evaluation_read(req: EvaluationReadReq) → EvaluationReadRes

`méthode` `evaluation_run_create`

evaluation_run_create(req: EvaluationRunCreateReq) → EvaluationRunCreateRes

`méthode` `evaluation_run_delete`

evaluation_run_delete(req: EvaluationRunDeleteReq) → EvaluationRunDeleteRes

`méthode` `evaluation_run_finish`

evaluation_run_finish(req: EvaluationRunFinishReq) → EvaluationRunFinishRes

`méthode` `evaluation_run_list`

evaluation_run_list(req: EvaluationRunListReq) → Iterator[EvaluationRunReadRes]

`méthode` `evaluation_run_read`

evaluation_run_read(req: EvaluationRunReadReq) → EvaluationRunReadRes

`méthode` `evaluation_status`

evaluation_status(req: EvaluationStatusReq) → EvaluationStatusRes

`méthode` `feedback_aggregate`

feedback_aggregate(req: FeedbackAggregateReq) → FeedbackAggregateRes

`méthode` `feedback_create`

feedback_create(req: FeedbackCreateReq) → FeedbackCreateRes

`méthode` `feedback_create_batch`

feedback_create_batch(req: FeedbackCreateBatchReq) → FeedbackCreateBatchRes

`méthode` `feedback_payload_schema`

feedback_payload_schema(
    req: FeedbackPayloadSchemaReq
) → FeedbackPayloadSchemaRes

`méthode` `feedback_purge`

feedback_purge(req: FeedbackPurgeReq) → FeedbackPurgeRes

`méthode` `feedback_query`

feedback_query(req: FeedbackQueryReq) → FeedbackQueryRes

`méthode` `feedback_replace`

feedback_replace(req: FeedbackReplaceReq) → FeedbackReplaceRes

`méthode` `feedback_stats`

feedback_stats(req: FeedbackStatsReq) → FeedbackStatsRes

`méthode` `file_content_read`

file_content_read(req: FileContentReadReq) → FileContentReadRes

`méthode` `file_create`

file_create(req: FileCreateReq) → FileCreateRes

`méthode` `files_stats`

files_stats(req: FilesStatsReq) → FilesStatsRes

`méthode` `genai_otel_export`

genai_otel_export(req: GenAIOTelExportReq) → GenAIOTelExportRes

`méthode` `image_create`

image_create(req: ImageGenerationCreateReq) → ImageGenerationCreateRes

`méthode` `model_create`

model_create(req: ModelCreateReq) → ModelCreateRes

`méthode` `model_delete`

model_delete(req: ModelDeleteReq) → ModelDeleteRes

`méthode` `model_list`

model_list(req: ModelListReq) → Iterator[ModelReadRes]

`méthode` `model_read`

model_read(req: ModelReadReq) → ModelReadRes

`méthode` `obj_add_tags`

obj_add_tags(req: ObjAddTagsReq) → ObjAddTagsRes

`méthode` `obj_create`

obj_create(req: ObjCreateReq) → ObjCreateRes

`méthode` `obj_delete`

obj_delete(req: ObjDeleteReq) → ObjDeleteRes

`méthode` `obj_read`

obj_read(req: ObjReadReq) → ObjReadRes

`méthode` `obj_remove_aliases`

obj_remove_aliases(req: ObjRemoveAliasesReq) → ObjRemoveAliasesRes

`méthode` `obj_remove_tags`

obj_remove_tags(req: ObjRemoveTagsReq) → ObjRemoveTagsRes

`méthode` `obj_set_aliases`

obj_set_aliases(req: ObjSetAliasesReq) → ObjSetAliasesRes

`méthode` `objs_query`

objs_query(req: ObjQueryReq) → ObjQueryRes

`méthode` `op_create`

op_create(req: OpCreateReq) → OpCreateRes

`méthode` `op_delete`

op_delete(req: OpDeleteReq) → OpDeleteRes

`méthode` `op_list`

op_list(req: OpListReq) → Iterator[OpReadRes]

`méthode` `op_read`

op_read(req: OpReadReq) → OpReadRes

`méthode` `otel_export`

otel_export(req: OTelExportReq) → OTelExportRes

`méthode` `prediction_create`

prediction_create(req: PredictionCreateReq) → PredictionCreateRes

`méthode` `prediction_delete`

prediction_delete(req: PredictionDeleteReq) → PredictionDeleteRes

`méthode` `prediction_finish`

prediction_finish(req: PredictionFinishReq) → PredictionFinishRes

`méthode` `prediction_list`

prediction_list(req: PredictionListReq) → Iterator[PredictionReadRes]

`méthode` `prediction_read`

prediction_read(req: PredictionReadReq) → PredictionReadRes

`méthode` `project_stats`

project_stats(req: ProjectStatsReq) → ProjectStatsRes

`méthode` `project_ttl_settings_read`

project_ttl_settings_read(
    req: ProjectTTLSettingsReadReq
) → ProjectTTLSettingsReadRes

`méthode` `project_ttl_settings_update`

project_ttl_settings_update(
    req: ProjectTTLSettingsUpdateReq
) → ProjectTTLSettingsUpdateRes

`méthode` `refs_read_batch`

refs_read_batch(req: RefsReadBatchReq) → RefsReadBatchRes

`méthode` `rescore`

rescore(req: RescoreReq) → RescoreRes

`méthode` `score_create`

score_create(req: ScoreCreateReq) → ScoreCreateRes

`méthode` `score_delete`

score_delete(req: ScoreDeleteReq) → ScoreDeleteRes

`méthode` `score_list`

score_list(req: ScoreListReq) → Iterator[ScoreReadRes]

`méthode` `score_read`

score_read(req: ScoreReadReq) → ScoreReadRes

`méthode` `scorer_create`

scorer_create(req: ScorerCreateReq) → ScorerCreateRes

`méthode` `scorer_delete`

scorer_delete(req: ScorerDeleteReq) → ScorerDeleteRes

`méthode` `scorer_list`

scorer_list(req: ScorerListReq) → Iterator[ScorerReadRes]

`méthode` `scorer_read`

scorer_read(req: ScorerReadReq) → ScorerReadRes

`méthode` `source_datasets_query`

source_datasets_query(req: SourceDatasetsQueryReq) → SourceDatasetsQueryRes

`méthode` `table_create`

table_create(req: TableCreateReq) → TableCreateRes

`méthode` `table_create_from_digests`

table_create_from_digests(
    req: TableCreateFromDigestsReq
) → TableCreateFromDigestsRes

`méthode` `table_query`

table_query(req: TableQueryReq) → TableQueryRes

`méthode` `table_query_stats`

table_query_stats(req: TableQueryStatsReq) → TableQueryStatsRes

`méthode` `table_query_stats_batch`

table_query_stats_batch(req: TableQueryStatsBatchReq) → TableQueryStatsBatchRes

`méthode` `table_query_stream`

table_query_stream(req: TableQueryReq) → Iterator[TableRowSchema]

`méthode` `table_update`

table_update(req: TableUpdateReq) → TableUpdateRes

`méthode` `tags_list`

tags_list(req: TagsListReq) → TagsListRes

`méthode` `threads_query_stream`

threads_query_stream(req: ThreadsQueryReq) → Iterator[ThreadSchema]

`méthode` `trace_usage`

trace_usage(req: 'TraceUsageReq') → TraceUsageRes

`classe` `GenAISpanRef`

Champs Pydantic :

trace_id: <class 'str'>
span_id: <class 'str'>

`classe` `ImageGenerationCreateReq`

Champs Pydantic :

project_id: <class 'str'>
inputs: <class 'ImageGenerationRequestInputs'>
wb_user_id: str | None
track_llm_call: bool | None

`classe` `ImageGenerationCreateRes`

Champs Pydantic :

response: dict[str, typing.Any]
weave_call_id: str | None

`classe` `ImageGenerationRequestInputs`

Champs Pydantic :

model: <class 'str'>
prompt: <class 'str'>
n: int | None

`classe` `LLMAggregatedUsage`

Métriques d’utilisation agrégées pour un LLM donné. Champs Pydantic :

requests: <class 'int'>
prompt_tokens: <class 'int'>
completion_tokens: <class 'int'>
total_tokens: <class 'int'>
cache_read_input_tokens: <class 'int'>
cache_creation_input_tokens: <class 'int'>
prompt_tokens_total_cost: float | None
completion_tokens_total_cost: float | None
cache_read_input_tokens_total_cost: float | None
cache_creation_input_tokens_total_cost: float | None

`classe` `LLMCostSchema`

`classe` `LLMUsageSchema`

`classe` `ModelCreateBody`

Champs Pydantic :

name: <class 'str'>
description: str | None
source_code: <class 'str'>
attributes: dict[str, typing.Any] | None

`classe` `ModelCreateReq`

Champs Pydantic :

name: <class 'str'>
description: str | None
source_code: <class 'str'>
attributes: dict[str, typing.Any] | None
project_id: <class 'str'>
wb_user_id: str | None

`classe` `ModelCreateRes`

Champs Pydantic :

digest: <class 'str'>
object_id: <class 'str'>
version_index: <class 'int'>
model_ref: <class 'str'>

`classe` `ModelDeleteReq`

champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digests: list[str] | None
wb_user_id: str | None

`classe` `ModelDeleteRes`

Champs Pydantic :

num_deleted: <class 'int'>

`classe` `ModelListReq`

Champs Pydantic :

project_id: <class 'str'>
limit: int | None
offset: int | None

`classe` `ModelReadReq`

champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digest: <class 'str'>

`classe` `ModelReadRes`

champs Pydantic :

object_id: <class 'str'>
digest: <class 'str'>
version_index: <class 'int'>
created_at: <class 'datetime.datetime'>
name: <class 'str'>
description: str | None
source_code: <class 'str'>
attributes: dict[str, typing.Any] | None

`classe` `OTelExportReq`

Champs Pydantic :

processed_spans: list[ProcessedResourceSpans]
project_id: <class 'str'>
wb_user_id: str | None

`classe` `OTelExportRes`

Champs Pydantic :

partial_success: ExportTracePartialSuccess | None

`classe` `ObjAddTagsReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digest: <class 'str'>
tags: list[str]
wb_user_id: str | None

`méthode` `validate_tags`

validate_tags() → ObjAddTagsReq

`classe` `ObjAddTagsRes`

`classe` `ObjCreateReq`

Champs Pydantic :

obj: <class 'ObjSchemaForInsert'>

`classe` `ObjCreateRes`

Champs Pydantic :

digest: <class 'str'>
object_id: str | None

`classe` `ObjDeleteReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digests: list[str] | None

`classe` `ObjDeleteRes`

Champs Pydantic :

num_deleted: <class 'int'>

`classe` `ObjQueryReq`

Champs Pydantic :

project_id: <class 'str'>
filter: ObjectVersionFilter | None
limit: int | None
offset: int | None
sort_by: list[weave.trace_server.common_interface.SortBy] | None
metadata_only: bool | None
include_storage_size: bool | None
include_tags_and_aliases: bool | None

`classe` `ObjQueryRes`

champs Pydantic :

objs: list[ObjSchema]

`classe` `ObjReadReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digest: <class 'str'>
metadata_only: bool | None
include_tags_and_aliases: bool | None

`classe` `ObjReadRes`

Champs Pydantic :

obj: <class 'ObjSchema'>

`classe` `ObjRemoveAliasesReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
aliases: list[str]
wb_user_id: str | None

`méthode` `validate_aliases`

validate_aliases() → ObjRemoveAliasesReq

`classe` `ObjRemoveAliasesRes`

`classe` `ObjRemoveTagsReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digest: <class 'str'>
tags: list[str]
wb_user_id: str | None

`classe` `ObjRemoveTagsRes`

`classe` `ObjSchema`

champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
created_at: <class 'datetime.datetime'>
deleted_at: datetime.datetime | None
digest: <class 'str'>
version_index: <class 'int'>
is_latest: <class 'int'>
kind: <class 'str'>
base_object_class: str | None
leaf_object_class: str | None
val: typing.Any
wb_user_id: str | None
size_bytes: int | None
tags: list[str] | None
aliases: list[str] | None

`classe` `ObjSchemaForInsert`

champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
val: typing.Any
builtin_object_class: str | None
set_base_object_class: str | None
expected_digest: str | None
wb_user_id: str | None

`méthode` `model_post_init`

model_post_init(context: Any) → None

`classe` `ObjSetAliasesReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digest: <class 'str'>
aliases: list[str]
wb_user_id: str | None

`méthode` `validate_aliases`

validate_aliases() → ObjSetAliasesReq

`classe` `ObjSetAliasesRes`

`classe` `ObjectInterface`

Points de terminaison de l’API des objets pour Trace Server. Ce protocole contient des API de gestion d’objets qui fournissent des interfaces plus claires et plus conformes aux principes REST. Les implémentations doivent prendre en charge à la fois ce protocole et TraceServerInterface afin de maintenir la rétrocompatibilité.

`méthode` `call_end_v2`

call_end_v2(req: CallEndV2Req) → CallEndV2Res

`méthode` `call_start_v2`

call_start_v2(req: CallStartV2Req) → CallStartV2Res

`méthode` `calls_complete`

calls_complete(req: CallsUpsertCompleteReq) → CallsUpsertCompleteRes

`méthode` `dataset_create`

dataset_create(req: DatasetCreateReq) → DatasetCreateRes

`méthode` `dataset_delete`

dataset_delete(req: DatasetDeleteReq) → DatasetDeleteRes

`méthode` `dataset_list`

dataset_list(req: DatasetListReq) → Iterator[DatasetReadRes]

`méthode` `dataset_read`

dataset_read(req: DatasetReadReq) → DatasetReadRes

`méthode` `eval_results_query`

eval_results_query(req: EvalResultsQueryReq) → EvalResultsQueryRes

`méthode` `evaluation_create`

evaluation_create(req: EvaluationCreateReq) → EvaluationCreateRes

`méthode` `evaluation_delete`

evaluation_delete(req: EvaluationDeleteReq) → EvaluationDeleteRes

`méthode` `evaluation_list`

evaluation_list(req: EvaluationListReq) → Iterator[EvaluationReadRes]

`méthode` `evaluation_read`

evaluation_read(req: EvaluationReadReq) → EvaluationReadRes

`méthode` `evaluation_run_create`

evaluation_run_create(req: EvaluationRunCreateReq) → EvaluationRunCreateRes

`méthode` `evaluation_run_delete`

evaluation_run_delete(req: EvaluationRunDeleteReq) → EvaluationRunDeleteRes

`méthode` `evaluation_run_finish`

evaluation_run_finish(req: EvaluationRunFinishReq) → EvaluationRunFinishRes

`méthode` `evaluation_run_list`

evaluation_run_list(req: EvaluationRunListReq) → Iterator[EvaluationRunReadRes]

`méthode` `evaluation_run_read`

evaluation_run_read(req: EvaluationRunReadReq) → EvaluationRunReadRes

`méthode` `model_create`

model_create(req: ModelCreateReq) → ModelCreateRes

`méthode` `model_delete`

model_delete(req: ModelDeleteReq) → ModelDeleteRes

`méthode` `model_list`

model_list(req: ModelListReq) → Iterator[ModelReadRes]

`méthode` `model_read`

model_read(req: ModelReadReq) → ModelReadRes

`méthode` `op_create`

op_create(req: OpCreateReq) → OpCreateRes

`méthode` `op_delete`

op_delete(req: OpDeleteReq) → OpDeleteRes

`méthode` `op_list`

op_list(req: OpListReq) → Iterator[OpReadRes]

`méthode` `op_read`

op_read(req: OpReadReq) → OpReadRes

`méthode` `prediction_create`

prediction_create(req: PredictionCreateReq) → PredictionCreateRes

`méthode` `prediction_delete`

prediction_delete(req: PredictionDeleteReq) → PredictionDeleteRes

`méthode` `prediction_finish`

prediction_finish(req: PredictionFinishReq) → PredictionFinishRes

`méthode` `prediction_list`

prediction_list(req: PredictionListReq) → Iterator[PredictionReadRes]

`méthode` `prediction_read`

prediction_read(req: PredictionReadReq) → PredictionReadRes

`méthode` `score_create`

score_create(req: ScoreCreateReq) → ScoreCreateRes

`méthode` `score_delete`

score_delete(req: ScoreDeleteReq) → ScoreDeleteRes

`méthode` `score_list`

score_list(req: ScoreListReq) → Iterator[ScoreReadRes]

`méthode` `score_read`

score_read(req: ScoreReadReq) → ScoreReadRes

`méthode` `scorer_create`

scorer_create(req: ScorerCreateReq) → ScorerCreateRes

`méthode` `scorer_delete`

scorer_delete(req: ScorerDeleteReq) → ScorerDeleteRes

`méthode` `scorer_list`

scorer_list(req: ScorerListReq) → Iterator[ScorerReadRes]

`méthode` `scorer_read`

scorer_read(req: ScorerReadReq) → ScorerReadRes

`classe` `ObjectVersionFilter`

Champs Pydantic :

base_object_classes: list[str] | None
exclude_base_object_classes: list[str] | None
leaf_object_classes: list[str] | None
object_ids: list[str] | None
is_op: bool | None
latest_only: bool | None
tags: list[str] | None
aliases: list[str] | None

`classe` `OpCreateBody`

Corps de requête pour créer un objet Op via l’API REST. Ce modèle exclut project_id, car il est fourni par le chemin de l’URL dans les endpoints REST. champs Pydantic :

name: str | None
source_code: str | None

`classe` `OpCreateReq`

Modèle de requête pour créer un objet Op. Étend OpCreateBody en y ajoutant project_id pour un usage interne de l’API. Champs Pydantic :

name: str | None
source_code: str | None
project_id: <class 'str'>
wb_user_id: str | None

`classe` `OpCreateRes`

Modèle de réponse pour créer un objet Op. Champs Pydantic :

digest: <class 'str'>
object_id: <class 'str'>
version_index: <class 'int'>

`classe` `OpDeleteReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digests: list[str] | None
wb_user_id: str | None

`classe` `OpDeleteRes`

champs Pydantic :

num_deleted: <class 'int'>

`classe` `OpListReq`

Champs Pydantic :

project_id: <class 'str'>
limit: int | None
offset: int | None
wb_user_id: str | None

`classe` `OpReadReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digest: <class 'str'>
wb_user_id: str | None

`classe` `OpReadRes`

Modèle de réponse pour lire un objet Op. Le champ code contient le code source réel de l’op. Champs Pydantic :

object_id: <class 'str'>
digest: <class 'str'>
version_index: <class 'int'>
created_at: <class 'datetime.datetime'>
code: <class 'str'>

`classe` `PredictionCreateBody`

Corps de requête pour créer une prédiction via l’API REST. Ce modèle exclut project_id, car il est récupéré à partir du chemin d’URL des endpoints REST. champs Pydantic :

model: <class 'str'>
inputs: dict[str, typing.Any]
output: typing.Any
evaluation_run_id: str | None
genai_span_ref: list[GenAISpanRef] | None

`classe` `PredictionCreateReq`

Modèle de requête pour créer une prédiction. Étend PredictionCreateBody en ajoutant project_id pour une utilisation interne de l’API. Champs Pydantic :

model: <class 'str'>
inputs: dict[str, typing.Any]
output: typing.Any
evaluation_run_id: str | None
genai_span_ref: list[GenAISpanRef] | None
project_id: <class 'str'>
wb_user_id: str | None

`classe` `PredictionCreateRes`

Champs Pydantic :

prediction_id: <class 'str'>

`classe` `PredictionDeleteReq`

Champs Pydantic :

project_id: <class 'str'>
prediction_ids: list[str]
wb_user_id: str | None

`classe` `PredictionDeleteRes`

champs Pydantic :

num_deleted: <class 'int'>

`classe` `PredictionFinishReq`

champs Pydantic :

project_id: <class 'str'>
prediction_id: <class 'str'>
wb_user_id: str | None

`classe` `PredictionFinishRes`

Champs Pydantic :

success: <class 'bool'>

`classe` `PredictionListReq`

Champs Pydantic :

project_id: <class 'str'>
evaluation_run_id: str | None
limit: int | None
offset: int | None
wb_user_id: str | None

`classe` `PredictionListRes`

Champs Pydantic :

predictions: list[PredictionReadRes]

`classe` `PredictionReadReq`

Champs Pydantic :

project_id: <class 'str'>
prediction_id: <class 'str'>
wb_user_id: str | None

`classe` `PredictionReadRes`

Champs Pydantic :

prediction_id: <class 'str'>
model: <class 'str'>
inputs: dict[str, typing.Any]
output: typing.Any
evaluation_run_id: str | None
wb_user_id: str | None

`classe` `ProcessedResourceSpans`

Champs Pydantic :

entity: <class 'str'>
project: <class 'str'>
run_id: str | None
resource_spans: <class 'opentelemetry.proto.trace.v1.trace_pb2.ResourceSpans'>

`classe` `ProjectStatsReq`

Champs Pydantic :

project_id: <class 'str'>
include_trace_storage_size: bool | None
include_object_storage_size: bool | None
include_table_storage_size: bool | None
include_file_storage_size: bool | None

`classe` `ProjectStatsRes`

Champs Pydantic :

trace_storage_size_bytes: <class 'int'>
objects_storage_size_bytes: <class 'int'>
tables_storage_size_bytes: <class 'int'>
files_storage_size_bytes: <class 'int'>

`classe` `ProjectTTLSettingsReadReq`

Champs Pydantic :

project_id: <class 'str'>

`classe` `ProjectTTLSettingsReadRes`

champs Pydantic :

retention_days: int | None

`classe` `ProjectTTLSettingsUpdateReq`

Champs Pydantic :

project_id: <class 'str'>
retention_days: int | None
wb_user_id: str | None

`classe` `ProjectTTLSettingsUpdateRes`

champs Pydantic :

retention_days: int | None

`classe` `RefsReadBatchReq`

Champs Pydantic :

refs : list[str]

`classe` `RefsReadBatchRes`

Champs Pydantic :

vals: list[typing.Any]

`classe` `RescoreBody`

Corps de requête pour le recalcul des scores via l’API REST (hors champs définis par le serveur). Champs Pydantic :

source_evaluation_run_id : <class 'str'>
scorer_refs : list[str]

`classe` `RescoreReq`

Requête complète de recalcul des scores, y compris les champs définis par le serveur. Champs Pydantic :

source_evaluation_run_id: <class 'str'>
scorer_refs: list[str]
project_id: <class 'str'>
wb_user_id: str | None

`classe` `RescoreRes`

Réponse à une requête de recalcul des scores. Champs Pydantic :

call_id: <class 'str'>
evaluation_run_id: <class 'str'>

`classe` `RescoringArgs`

Arguments d’un job de recalcul des scores envoyé au worker evaluate-model. Diffère de EvaluateModelArgs : aucun modèle n’est chargé et aucune prédiction n’est exécutée. Seuls les scorer(s) sont appliqués aux prédictions existantes de source_evaluation_run_id. Champs Pydantic :

job_type: typing.Literal['rescore']
project_id: <class 'str'>
source_evaluation_run_id: <class 'str'>
scorer_refs: list[str]
wb_user_id: str | None
new_evaluation_run_id: <class 'str'>

`classe` `ScoreCreateBody`

corps de requête pour créer un score via l’API REST. Ce modèle exclut project_id, car il provient du chemin de l’URL dans les endpoints REST. Champs Pydantic :

prediction_id: <class 'str'>
scorer: <class 'str'>
value: typing.Any
evaluation_run_id: str | None

`classe` `ScoreCreateReq`

Modèle de requête pour créer un score. Étend ScoreCreateBody en y ajoutant project_id pour l’usage interne de l’API. Champs Pydantic :

prediction_id: <class 'str'>
scorer: <class 'str'>
value: typing.Any
evaluation_run_id: str | None
project_id: <class 'str'>
wb_user_id: str | None

`classe` `ScoreCreateRes`

Champs Pydantic :

score_id: <class 'str'>

`classe` `ScoreDeleteReq`

Champs Pydantic :

project_id: <class 'str'>
score_ids: list[str]
wb_user_id: str | None

`classe` `ScoreDeleteRes`

Champs Pydantic :

num_deleted: <class 'int'>

`classe` `ScoreListReq`

Champs Pydantic :

project_id: <class 'str'>
evaluation_run_id: str | None
limit: int | None
offset: int | None
wb_user_id: str | None

`classe` `ScoreReadReq`

champs Pydantic :

project_id: <class 'str'>
score_id: <class 'str'>
wb_user_id: str | None

`classe` `ScoreReadRes`

Champs Pydantic :

score_id: <class 'str'>
scorer: <class 'str'>
value: typing.Any
evaluation_run_id: str | None
wb_user_id: str | None

`classe` `ScorerCreateBody`

Champs Pydantic :

name: <class 'str'>
description: str | None
op_source_code: <class 'str'>

`classe` `ScorerCreateReq`

Champs Pydantic :

name: <class 'str'>
description: str | None
op_source_code: <class 'str'>
project_id: <class 'str'>
wb_user_id: str | None

`classe` `ScorerCreateRes`

Champs Pydantic :

digest: <class 'str'>
object_id: <class 'str'>
version_index: <class 'int'>
scorer: <class 'str'>

`classe` `ScorerDeleteReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digests: list[str] | None
wb_user_id: str | None

`classe` `ScorerDeleteRes`

champs Pydantic :

num_deleted: <class 'int'>

`classe` `ScorerListReq`

Champs Pydantic :

project_id: <class 'str'>
limit: int | None
offset: int | None
wb_user_id: str | None

`classe` `ScorerReadReq`

Champs Pydantic :

project_id: <class 'str'>
object_id: <class 'str'>
digest: <class 'str'>
wb_user_id: str | None

`classe` `ScorerReadRes`

champs Pydantic :

object_id: <class 'str'>
digest: <class 'str'>
version_index: <class 'int'>
created_at: <class 'datetime.datetime'>
name: <class 'str'>
description: str | None
score_op: <class 'str'>

`classe` `SourceDatasetMembership`

Appartenance d’une paire unique (source, jeu de données) dans la requête inverse. Champs Pydantic :

source_kind: <enum 'SourceKind'>
source_id: <class 'str'>
source_trace_id: <class 'str'>
dataset_object_id: <class 'str'>
row_digests: list[str]
row_digests_truncated: <class 'bool'>
row_digests_total_count: <class 'int'>
first_seen_at: <class 'datetime.datetime'>

`classe` `SourceDatasetsQueryReq`

Requête inverse : sources -> jeux de données. Champs Pydantic :

project_id: <class 'str'>
sources: list[SourceRef]
include_deleted: <class 'bool'>
wb_user_id: str | None

`classe` `SourceDatasetsQueryRes`

Réponse à la requête inverse sources -> jeux de données. Champs Pydantic :

memberships: list[SourceDatasetMembership]

`classe` `SourceKind`

`classe` `SourceRef`

Référence à une source d’origine (un appel, un span d’agent ou une conversation). champs Pydantic :

source_kind: <enum 'SourceKind'>
source_id: <class 'str'>
source_trace_id: <class 'str'>

`classe` `StartedCallSchemaForInsert`

champs Pydantic :

project_id: <class 'str'>
id: str | None
op_name: <class 'str'>
display_name: str | None
trace_id: str | None
parent_id: str | None
thread_id: str | None
turn_id: str | None
started_at: <class 'datetime.datetime'>
attributes: dict[str, typing.Any]
inputs: dict[str, typing.Any]
otel_dump: dict[str, typing.Any] | None
wb_user_id: str | None
wb_run_id: str | None
wb_run_step: int | None

`classe` `SummaryInsertMap`

`classe` `SummaryMap`

`classe` `TableAppendSpec`

Champs Pydantic :

append: <class 'TableAppendSpecPayload'>

`classe` `TableAppendSpecPayload`

Champs Pydantic :

row : dict[str, typing.Any]

`classe` `TableCreateFromDigestsReq`

Champs Pydantic :

project_id: <class 'str'>
row_digests: list[str]
expected_digest: str | None

`classe` `TableCreateFromDigestsRes`

champs Pydantic :

digest : <class 'str'>

`classe` `TableCreateReq`

champs Pydantic :

table: <class 'TableSchemaForInsert'>

`classe` `TableCreateRes`

Champs Pydantic :

digest: <class 'str'>
row_digests: list[str]

`classe` `TableInsertSpec`

champs Pydantic :

insert: <class 'TableInsertSpecPayload'>

`classe` `TableInsertSpecPayload`

champs Pydantic :

index: <class 'int'>
row: dict[str, typing.Any]

`classe` `TablePopSpec`

Champs Pydantic :

pop: <class 'TablePopSpecPayload'>

`classe` `TablePopSpecPayload`

Champs Pydantic :

index: <class 'int'>

`classe` `TableQueryReq`

Champs Pydantic :

project_id: <class 'str'>
digest: <class 'str'>
filter: TableRowFilter | None
limit: int | None
offset: int | None
sort_by: list[weave.trace_server.common_interface.SortBy] | None

`classe` `TableQueryRes`

Champs Pydantic :

rows: list[TableRowSchema]

`classe` `TableQueryStatsBatchReq`

Champs Pydantic :

project_id: <class 'str'>
digests: list[str] | None
include_storage_size: bool | None

`classe` `TableQueryStatsBatchRes`

Champs Pydantic :

tables: list[TableStatsRow]

`classe` `TableQueryStatsReq`

Champs Pydantic :

project_id: <class 'str'>
digest: <class 'str'>

`classe` `TableQueryStatsRes`

champs Pydantic :

count: <class 'int'>

`classe` `TableRowFilter`

Champs Pydantic :

row_digests: list[str] | None

`classe` `TableRowSchema`

champs Pydantic :

digest: <class 'str'>
val: typing.Any
original_index: int | None

`classe` `TableSchemaForInsert`

champs Pydantic :

project_id: <class 'str'>
rows: list[dict[str, typing.Any]]
expected_digest: str | None

`classe` `TableStatsRow`

Champs Pydantic :

count: <class 'int'>
digest: <class 'str'>
storage_size_bytes: int | None

`classe` `TableUpdateReq`

Champs Pydantic :

project_id: <class 'str'>
base_digest: <class 'str'>
updates: list[TableAppendSpec | TablePopSpec | TableInsertSpec]

`classe` `TableUpdateRes`

Champs Pydantic :

digest: <class 'str'>
updated_row_digests: list[str]

`classe` `TagsListReq`

Champs Pydantic :

project_id: <class 'str'>
wb_user_id: str | None

`classe` `TagsListRes`

Champs Pydantic :

tags : list[str]

`classe` `ThreadSchema`

Champs Pydantic :

thread_id: <class 'str'>
turn_count: <class 'int'>
start_time: <class 'datetime.datetime'>
last_updated: <class 'datetime.datetime'>
first_turn_id: str | None
last_turn_id: str | None
p50_turn_duration_ms: float | None
p99_turn_duration_ms: float | None

`classe` `ThreadsQueryFilter`

Champs Pydantic :

after_datetime: datetime.datetime | None
before_datetime: datetime.datetime | None
thread_ids: list[str] | None

`classe` `ThreadsQueryReq`

Interroge les threads avec des statistiques agrégées calculées uniquement à partir des appels de tour. Les appels de tour sont les enfants directs des contextes de thread (où call.id == turn_id). Cela fournit des statistiques pertinentes à l’échelle de la conversation, plutôt que d’inclure tous les détails d’implémentation imbriqués. Champs Pydantic :

project_id: <class 'str'>
filter: ThreadsQueryFilter | None
limit: int | None
offset: int | None
sort_by: list[weave.trace_server.common_interface.SortBy] | None

`classe` `TraceServerInterface`

`méthode` `agent_agents_query`

agent_agents_query(req: AgentsQueryReq) → AgentsQueryRes

`méthode` `agent_conversation_chat`

agent_conversation_chat(
    req: AgentConversationChatReq
) → AgentConversationChatRes

`méthode` `agent_conversation_spans`

agent_conversation_spans(
    req: AgentConversationSpansReq
) → AgentConversationSpansRes

`méthode` `agent_custom_attrs_schema`

agent_custom_attrs_schema(
    req: AgentCustomAttrsSchemaReq
) → AgentCustomAttrsSchemaRes

`méthode` `agent_search`

agent_search(req: AgentSearchReq) → AgentSearchRes

`méthode` `agent_spans_query`

agent_spans_query(req: AgentSpansQueryReq) → AgentSpansQueryRes

`méthode` `agent_spans_stats`

agent_spans_stats(req: AgentSpanStatsReq) → AgentSpanStatsRes

`méthode` `agent_traces_chat`

agent_traces_chat(req: AgentTraceChatReq) → AgentTraceChatRes

`méthode` `agent_versions_query`

agent_versions_query(req: AgentVersionsQueryReq) → AgentVersionsQueryRes

`méthode` `aliases_list`

aliases_list(req: AliasesListReq) → AliasesListRes

`méthode` `annotation_queue_add_calls`

annotation_queue_add_calls(
    req: AnnotationQueueAddCallsReq
) → AnnotationQueueAddCallsRes

`méthode` `annotation_queue_create`

annotation_queue_create(
    req: AnnotationQueueCreateReq
) → AnnotationQueueCreateRes

`méthode` `annotation_queue_delete`

annotation_queue_delete(
    req: AnnotationQueueDeleteReq
) → AnnotationQueueDeleteRes

`méthode` `annotation_queue_items_query`

annotation_queue_items_query(
    req: AnnotationQueueItemsQueryReq
) → AnnotationQueueItemsQueryRes

`méthode` `annotation_queue_read`

annotation_queue_read(req: AnnotationQueueReadReq) → AnnotationQueueReadRes

`méthode` `annotation_queue_update`

annotation_queue_update(
    req: AnnotationQueueUpdateReq
) → AnnotationQueueUpdateRes

`méthode` `annotation_queues_query_stream`

annotation_queues_query_stream(
    req: AnnotationQueuesQueryReq
) → Iterator[AnnotationQueueSchema]

`méthode` `annotation_queues_stats`

annotation_queues_stats(
    req: AnnotationQueuesStatsReq
) → AnnotationQueuesStatsRes

`méthode` `annotator_queue_items_progress_update`

annotator_queue_items_progress_update(
    req: AnnotatorQueueItemsProgressUpdateReq
) → AnnotatorQueueItemsProgressUpdateRes

`méthode` `call_end`

call_end(req: CallEndReq) → CallEndRes

`méthode` `call_read`

call_read(req: CallReadReq) → CallReadRes

`méthode` `call_start`

call_start(req: CallStartReq) → CallStartRes

`méthode` `call_start_batch`

call_start_batch(req: CallCreateBatchReq) → CallCreateBatchRes

`méthode` `call_stats`

call_stats(req: 'CallStatsReq') → CallStatsRes

`méthode` `call_update`

call_update(req: CallUpdateReq) → CallUpdateRes

`méthode` `calls_delete`

calls_delete(req: CallsDeleteReq) → CallsDeleteRes

`méthode` `calls_query`

calls_query(req: CallsQueryReq) → CallsQueryRes

`méthode` `calls_query_stats`

calls_query_stats(req: CallsQueryStatsReq) → CallsQueryStatsRes

`méthode` `calls_query_stream`

calls_query_stream(req: CallsQueryReq) → Iterator[CallSchema]

`méthode` `calls_score`

calls_score(req: CallsScoreReq) → CallsScoreRes

`méthode` `calls_usage`

calls_usage(req: 'CallsUsageReq') → CallsUsageRes

`méthode` `completions_create`

completions_create(req: CompletionsCreateReq) → CompletionsCreateRes

`méthode` `completions_create_stream`

completions_create_stream(req: CompletionsCreateReq) → Iterator[dict[str, Any]]

`méthode` `cost_create`

cost_create(req: CostCreateReq) → CostCreateRes

`méthode` `cost_purge`

cost_purge(req: CostPurgeReq) → CostPurgeRes

`méthode` `cost_query`

cost_query(req: CostQueryReq) → CostQueryRes

`méthode` `dataset_sources_link`

dataset_sources_link(req: DatasetSourcesLinkReq) → DatasetSourcesLinkRes

`méthode` `dataset_sources_link_delete`

dataset_sources_link_delete(
    req: DatasetSourcesLinkDeleteReq
) → DatasetSourcesLinkDeleteRes

`méthode` `dataset_sources_query`

dataset_sources_query(req: DatasetSourcesQueryReq) → DatasetSourcesQueryRes

`méthode` `evaluate_model`

evaluate_model(req: EvaluateModelReq) → EvaluateModelRes

`méthode` `evaluation_status`

evaluation_status(req: EvaluationStatusReq) → EvaluationStatusRes

`méthode` `feedback_aggregate`

feedback_aggregate(req: FeedbackAggregateReq) → FeedbackAggregateRes

`méthode` `feedback_create`

feedback_create(req: FeedbackCreateReq) → FeedbackCreateRes

`méthode` `feedback_create_batch`

feedback_create_batch(req: FeedbackCreateBatchReq) → FeedbackCreateBatchRes

`méthode` `feedback_payload_schema`

feedback_payload_schema(
    req: FeedbackPayloadSchemaReq
) → FeedbackPayloadSchemaRes

`méthode` `feedback_purge`

feedback_purge(req: FeedbackPurgeReq) → FeedbackPurgeRes

`méthode` `feedback_query`

feedback_query(req: FeedbackQueryReq) → FeedbackQueryRes

`méthode` `feedback_replace`

feedback_replace(req: FeedbackReplaceReq) → FeedbackReplaceRes

`méthode` `feedback_stats`

feedback_stats(req: FeedbackStatsReq) → FeedbackStatsRes

`méthode` `file_content_read`

file_content_read(req: FileContentReadReq) → FileContentReadRes

`méthode` `file_create`

file_create(req: FileCreateReq) → FileCreateRes

`méthode` `files_stats`

files_stats(req: FilesStatsReq) → FilesStatsRes

`méthode` `genai_otel_export`

genai_otel_export(req: GenAIOTelExportReq) → GenAIOTelExportRes

`méthode` `image_create`

image_create(req: ImageGenerationCreateReq) → ImageGenerationCreateRes

`méthode` `obj_add_tags`

obj_add_tags(req: ObjAddTagsReq) → ObjAddTagsRes

`méthode` `obj_create`

obj_create(req: ObjCreateReq) → ObjCreateRes

`méthode` `obj_delete`

obj_delete(req: ObjDeleteReq) → ObjDeleteRes

`méthode` `obj_read`

obj_read(req: ObjReadReq) → ObjReadRes

`méthode` `obj_remove_aliases`

obj_remove_aliases(req: ObjRemoveAliasesReq) → ObjRemoveAliasesRes

`méthode` `obj_remove_tags`

obj_remove_tags(req: ObjRemoveTagsReq) → ObjRemoveTagsRes

`méthode` `obj_set_aliases`

obj_set_aliases(req: ObjSetAliasesReq) → ObjSetAliasesRes

`méthode` `objs_query`

objs_query(req: ObjQueryReq) → ObjQueryRes

`méthode` `otel_export`

otel_export(req: OTelExportReq) → OTelExportRes

`méthode` `project_stats`

project_stats(req: ProjectStatsReq) → ProjectStatsRes

`méthode` `project_ttl_settings_read`

project_ttl_settings_read(
    req: ProjectTTLSettingsReadReq
) → ProjectTTLSettingsReadRes

`méthode` `project_ttl_settings_update`

project_ttl_settings_update(
    req: ProjectTTLSettingsUpdateReq
) → ProjectTTLSettingsUpdateRes

`méthode` `refs_read_batch`

refs_read_batch(req: RefsReadBatchReq) → RefsReadBatchRes

`méthode` `rescore`

rescore(req: RescoreReq) → RescoreRes

`méthode` `source_datasets_query`

source_datasets_query(req: SourceDatasetsQueryReq) → SourceDatasetsQueryRes

`méthode` `table_create`

table_create(req: TableCreateReq) → TableCreateRes

`méthode` `table_create_from_digests`

table_create_from_digests(
    req: TableCreateFromDigestsReq
) → TableCreateFromDigestsRes

`méthode` `table_query`

table_query(req: TableQueryReq) → TableQueryRes

`méthode` `table_query_stats`

table_query_stats(req: TableQueryStatsReq) → TableQueryStatsRes

`méthode` `table_query_stats_batch`

table_query_stats_batch(req: TableQueryStatsBatchReq) → TableQueryStatsBatchRes

`méthode` `table_query_stream`

table_query_stream(req: TableQueryReq) → Iterator[TableRowSchema]

`méthode` `table_update`

table_update(req: TableUpdateReq) → TableUpdateRes

`méthode` `tags_list`

tags_list(req: TagsListReq) → TagsListRes

`méthode` `threads_query_stream`

threads_query_stream(req: ThreadsQueryReq) → Iterator[ThreadSchema]

`méthode` `trace_usage`

trace_usage(req: 'TraceUsageReq') → TraceUsageRes

`class` `TraceStatus`

`class` `TraceUsageReq`

Requête permettant de calculer l’utilisation par appel pour une trace, avec cumul des descendants. Ce point de terminaison renvoie des métriques d’utilisation pour chaque appel de la trace. Les métriques de chaque appel incluent la somme de sa propre utilisation et de celle de tous ses descendants. Utilisez-le dans la Vue de trace lorsque vous souhaitez afficher des métriques cumulées par appel. Remarque : tous les appels correspondants sont chargés en mémoire pour l’agrégation. Pour les ensembles de résultats très volumineux (>10k appels), envisagez d’utiliser des filtres plus spécifiques ou une pagination au niveau de l’application. Champs Pydantic :

project_id: <class 'str'>
filter: CallsFilter | None
query: weave.trace_server.interface.query.Query | None
include_costs: <class 'bool'>
limit: <class 'int'>

`class` `TraceUsageRes`

Réponse contenant les métriques d’utilisation par appel (chacune inclut les contributions des descendants). Champs Pydantic :

call_usage: dict[str, dict[str, LLMAggregatedUsage]]
unfinished_call_ids: list[str]

`class` `UsageMetricSpec`

Spécification d’une métrique d’utilisation à agréger (groupée par modèle). Champs Pydantic :

metric: typing.Literal['input_tokens', 'output_tokens', 'total_tokens', 'cache_read_input_tokens', 'cache_creation_input_tokens', 'input_cost', 'output_cost', 'total_cost']
aggregations: list[AggregationType]
percentiles: list[float]

​Aperçu de l’API

​classe AggregationType

​classe AliasesListReq

​classe AliasesListRes

​classe AnnotationQueueAddCallsReq

​classe AnnotationQueueAddCallsRes

​classe AnnotationQueueCreateReq

​classe AnnotationQueueCreateRes

​classe AnnotationQueueDeleteReq

​classe AnnotationQueueDeleteRes

​classe AnnotationQueueItemSchema

​classe AnnotationQueueItemsQueryReq

​classe AnnotationQueueItemsQueryRes

​classe AnnotationQueueReadReq

​classe AnnotationQueueReadRes

​classe AnnotationQueueSchema

​classe AnnotationQueueStatsSchema

​classe AnnotationQueueUpdateReq

​classe AnnotationQueueUpdateRes

​classe AnnotationQueuesQueryReq

​classe AnnotationQueuesQueryRes

​classe AnnotationQueuesStatsReq

​classe AnnotationQueuesStatsRes

​classe AnnotatorQueueItemsProgressUpdateReq

​classe AnnotatorQueueItemsProgressUpdateRes

​classe CallBatchEndMode

​classe CallBatchStartMode

​classe CallCreateBatchReq

​classe CallCreateBatchRes

​classe CallEndReq

​classe CallEndRes

​classe CallEndV2Req

​classe CallEndV2Res

​classe CallMetricSpec

​classe CallReadReq

​classe CallReadRes

​classe CallSchema

​méthode serialize_typed_dicts

​classe CallStartReq

​classe CallStartRes

​classe CallStartV2Req

​classe CallStartV2Res

​classe CallStatsReq

​méthode validate_date_range

​classe CallStatsRes

​classe CallUpdateReq

​classe CallUpdateRes

​classe CallsDeleteReq

​classe CallsDeleteRes

​classe CallsFilter

​classe CallsQueryReq

​classe CallsQueryRes

​classe CallsQueryStatsReq

​classe CallsQueryStatsRes

​classe CallsScoreReq

​classe CallsScoreRes

​classe CallsUpsertCompleteReq

​classe CallsUpsertCompleteRes

​classe CallsUsageReq

​classe CallsUsageRes

​classe CompletedCallSchemaForInsert

​méthode serialize_typed_dicts

​classe CompletionsCreateReq

​classe CompletionsCreateRequestInputs

​classe CompletionsCreateRes

​classe CostCreateInput

​classe CostCreateReq

​classe CostCreateRes

​classe CostPurgeReq

​classe CostPurgeRes

​classe CostQueryOutput

​classe CostQueryReq

​classe CostQueryRes

​classe DatasetCreateBody

​classe DatasetCreateReq

​classe DatasetCreateRes

​classe DatasetDeleteReq

​classe DatasetDeleteRes

​classe DatasetListReq

​classe DatasetReadReq

Aperçu de l’API

`classe` `AggregationType`

`classe` `AliasesListReq`

`classe` `AliasesListRes`

`classe` `AnnotationQueueAddCallsReq`

`classe` `AnnotationQueueAddCallsRes`

`classe` `AnnotationQueueCreateReq`

`classe` `AnnotationQueueCreateRes`

`classe` `AnnotationQueueDeleteReq`

`classe` `AnnotationQueueDeleteRes`

`classe` `AnnotationQueueItemSchema`

`classe` `AnnotationQueueItemsQueryReq`

`classe` `AnnotationQueueItemsQueryRes`

`classe` `AnnotationQueueReadReq`

`classe` `AnnotationQueueReadRes`

`classe` `AnnotationQueueSchema`

`classe` `AnnotationQueueStatsSchema`

`classe` `AnnotationQueueUpdateReq`

`classe` `AnnotationQueueUpdateRes`

`classe` `AnnotationQueuesQueryReq`

`classe` `AnnotationQueuesQueryRes`

`classe` `AnnotationQueuesStatsReq`

`classe` `AnnotationQueuesStatsRes`

`classe` `AnnotatorQueueItemsProgressUpdateReq`

`classe` `AnnotatorQueueItemsProgressUpdateRes`

`classe` `CallBatchEndMode`

`classe` `CallBatchStartMode`

`classe` `CallCreateBatchReq`

`classe` `CallCreateBatchRes`

`classe` `CallEndReq`

`classe` `CallEndRes`

`classe` `CallEndV2Req`

`classe` `CallEndV2Res`

`classe` `CallMetricSpec`

`classe` `CallReadReq`

`classe` `CallReadRes`

`classe` `CallSchema`

`méthode` `serialize_typed_dicts`

`classe` `CallStartReq`

`classe` `CallStartRes`

`classe` `CallStartV2Req`

`classe` `CallStartV2Res`

`classe` `CallStatsReq`

`méthode` `validate_date_range`

`classe` `CallStatsRes`

`classe` `CallUpdateReq`

`classe` `CallUpdateRes`

`classe` `CallsDeleteReq`

`classe` `CallsDeleteRes`

`classe` `CallsFilter`

`classe` `CallsQueryReq`

`classe` `CallsQueryRes`

`classe` `CallsQueryStatsReq`

`classe` `CallsQueryStatsRes`

`classe` `CallsScoreReq`

`classe` `CallsScoreRes`

`classe` `CallsUpsertCompleteReq`

`classe` `CallsUpsertCompleteRes`

`classe` `CallsUsageReq`

`classe` `CallsUsageRes`

`classe` `CompletedCallSchemaForInsert`

`méthode` `serialize_typed_dicts`

`classe` `CompletionsCreateReq`

`classe` `CompletionsCreateRequestInputs`

`classe` `CompletionsCreateRes`

`classe` `CostCreateInput`

`classe` `CostCreateReq`

`classe` `CostCreateRes`

`classe` `CostPurgeReq`

`classe` `CostPurgeRes`

`classe` `CostQueryOutput`

`classe` `CostQueryReq`

`classe` `CostQueryRes`

`classe` `DatasetCreateBody`

`classe` `DatasetCreateReq`

`classe` `DatasetCreateRes`

`classe` `DatasetDeleteReq`

`classe` `DatasetDeleteRes`

`classe` `DatasetListReq`

`classe` `DatasetReadReq`