Skip to content

cayu.evals.result_presentation

Public Cayu API. Definitions are grouped by their public source module, and definitions from private implementation modules appear on their package’s page; adding a symbol does not move existing links. The import shown at each entry is a supported public namespace. Read the Python API overview and install the required extra.

from cayu.evals import EVAL_RESULT_PRESENTATION_MAX_BYTES
EVAL_RESULT_PRESENTATION_MAX_BYTES = 32 << 20

Release source · Runtime guarantees · Capability guides

from cayu.evals import EVAL_RESULT_PRESENTATION_SCHEMA_VERSION
EVAL_RESULT_PRESENTATION_SCHEMA_VERSION = 2

Release source · Runtime guarantees · Capability guides

from cayu.evals import eval_result_report_from_json

Parse one bounded explainable-result report without format guessing.

eval_result_report_from_json(source: str | bytes | bytearray) -> EvalResultReportV2

Release source · Runtime guarantees · Capability guides

from cayu.evals import EVAL_RESULT_REPORT_MAX_BYTES
EVAL_RESULT_REPORT_MAX_BYTES = 96 << 20

Release source · Runtime guarantees · Capability guides

from cayu.evals import EVAL_RESULT_REPORT_SCHEMA_VERSION
EVAL_RESULT_REPORT_SCHEMA_VERSION = 2

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalAssertionPresentationV1

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
assertion_id StrictStr required
assertion_revision StrictStr required
kind StrictStr required
category EvalAssertionCategory required
outcome PublishedOutcome required
score StrictFloat | None Field(default=None, ge=0.0, le=1.0)
model_judge PublishedModelJudgeDetail | None None
structured_judge EvalStructuredJudgePresentationV1 | None None
tool_json PublishedToolArgumentsContainDetail | PublishedToolResultContainsDetail | None None
process PublishedChildStatusDetail | PublishedProcessEventDetail | PublishedProcessEventsInOrderDetail | None None
structure PublishedWorkspaceFileDetail | PublishedArtifactDetail | None None
memory PublishedMemoryAttributionDetail | None None
EvalAssertionPresentationV1.copy_model_judge(value: object) -> object
EvalAssertionPresentationV1.copy_tool_json(value: object) -> object
EvalAssertionPresentationV1.copy_process(value: object) -> object
EvalAssertionPresentationV1.copy_structure(value: object) -> object
EvalAssertionPresentationV1.copy_memory(value: object) -> object
EvalAssertionPresentationV1.validate_assertion_id(value: str, info) -> str
EvalAssertionPresentationV1.validate_assertion_revision(value: str, info) -> str
EvalAssertionPresentationV1.validate_kind(value: str, info) -> str
EvalAssertionPresentationV1.validate_assertion() -> EvalAssertionPresentationV1

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalCasePresentationV1

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
case_id StrictStr required
case_revision StrictStr required
status PublishedStatus required
score StrictFloat | None Field(default=None, ge=0.0, le=1.0)
dimensions EvalResultOutcomeDimensionsV1 required
trials tuple[EvalTrialPresentationV1, ...] Field(min_length=1, max_length=EVAL_CORPUS_MAX_TRIALS)
EvalCasePresentationV1.validate_case_id(value: str, info) -> str
EvalCasePresentationV1.validate_case_revision(value: str, info) -> str
EvalCasePresentationV1.validate_trials_order(value: object, info) -> object
EvalCasePresentationV1.validate_case() -> EvalCasePresentationV1

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalCasePresentationV2

Case presentation with an explicit retained-trial reliability decision.

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
case_id StrictStr required
case_revision StrictStr required
status PublishedStatus required
score StrictFloat | None Field(default=None, ge=0.0, le=1.0)
dimensions EvalResultOutcomeDimensionsV1 required
reliability EvalCaseReliabilityV1 required
trials tuple[EvalTrialPresentationV1, ...] Field(min_length=1, max_length=EVAL_CORPUS_MAX_TRIALS)
EvalCasePresentationV2.validate_case_id(value: str, info) -> str
EvalCasePresentationV2.validate_case_revision(value: str, info) -> str
EvalCasePresentationV2.validate_trials_order(value: object, info) -> object
EvalCasePresentationV2.validate_case() -> EvalCasePresentationV2

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalResultOutcomeDimensionsV1

Orthogonal result states that must not be collapsed into one opaque status.

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
candidate EvalCandidateOutcome required
deterministic_assertions EvalAssertionDimension required
semantic_quality EvalAssertionDimension required
evaluator_health EvalEvaluatorHealth required
runtime EvalRuntimeOutcome required
evidence EvalEvidenceState required

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalResultPresentationV1

Canonical explainable projection for one immutable public eval result.

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
schema_version Literal[1] 1
result_revision StrictStr required
evaluation_revision StrictStr required
origin EvalResultOrigin required
target_key StrictStr required
application_release_id StrictStr required
app_manifest_fingerprint StrictStr required
corpus_revision StrictStr required
suite_id StrictStr required
suite_revision StrictStr required
evidence_policy_revision StrictStr required
pricing_profile_fingerprint StrictStr | None required
status PublishedStatus required
score StrictFloat | None Field(default=None, ge=0.0, le=1.0)
dimensions EvalResultOutcomeDimensionsV1 required
cases tuple[EvalCasePresentationV1, ...] Field(min_length=1, max_length=EVAL_CORPUS_MAX_CASES)
EvalResultPresentationV1.validate_schema_version_type(value: object) -> object
EvalResultPresentationV1.validate_revisions(value: str | None, info) -> str | None
EvalResultPresentationV1.validate_ids(value: str, info) -> str
EvalResultPresentationV1.validate_application_release_id(value: str, info) -> str

validate_app_manifest_fingerprint (method)

Section titled “validate_app_manifest_fingerprint (method)”
EvalResultPresentationV1.validate_app_manifest_fingerprint(value: str) -> str
EvalResultPresentationV1.validate_cases_order(value: object, info) -> object
EvalResultPresentationV1.validate_result_size() -> EvalResultPresentationV1

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalResultPresentationV2

Canonical V2 presentation with reliability policy and accepted exposure.

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
schema_version Literal[2] EVAL_RESULT_PRESENTATION_SCHEMA_VERSION
result_revision StrictStr required
evaluation_revision StrictStr required
origin EvalResultOrigin required
target_key StrictStr required
application_release_id StrictStr required
app_manifest_fingerprint StrictStr required
corpus_revision StrictStr required
suite_id StrictStr required
suite_revision StrictStr required
evidence_policy_revision StrictStr required
pricing_profile_fingerprint StrictStr | None required
trial_policy EvalSuiteTrialPolicyV1 required
accepted_exposure EvalSuiteRunExposureV1 | None None
status PublishedStatus required
score StrictFloat | None Field(default=None, ge=0.0, le=1.0)
dimensions EvalResultOutcomeDimensionsV1 required
cases tuple[EvalCasePresentationV2, ...] Field(min_length=1, max_length=EVAL_CORPUS_MAX_CASES)
EvalResultPresentationV2.validate_schema_version_type(value: object) -> object
EvalResultPresentationV2.validate_revisions(value: str | None, info) -> str | None
EvalResultPresentationV2.validate_ids(value: str, info) -> str
EvalResultPresentationV2.validate_application_release_id(value: str, info) -> str

validate_app_manifest_fingerprint (method)

Section titled “validate_app_manifest_fingerprint (method)”
EvalResultPresentationV2.validate_app_manifest_fingerprint(value: str) -> str
EvalResultPresentationV2.validate_cases_order(value: object, info) -> object
EvalResultPresentationV2.validate_result() -> EvalResultPresentationV2

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalResultReportV1

Portable report that binds immutable source evidence to its canonical view.

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
schema_version Literal[1] 1
record_type Literal['cayu.eval-result-report'] 'cayu.eval-result-report'
result CorpusExecutionResult | CapturedEvaluationResultV1 required
presentation EvalResultPresentationV1 required
EvalResultReportV1.validate_schema_version_type(value: object) -> object
EvalResultReportV1.copy_result(value: object) -> object
EvalResultReportV1.validate_report() -> EvalResultReportV1

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalResultReportV2

Current report binding immutable evidence to reliability presentation V2.

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
schema_version Literal[2] EVAL_RESULT_REPORT_SCHEMA_VERSION
record_type Literal['cayu.eval-result-report'] 'cayu.eval-result-report'
result CorpusExecutionResult | CapturedEvaluationResultV1 required
presentation EvalResultPresentationV2 required
EvalResultReportV2.validate_schema_version_type(value: object) -> object
EvalResultReportV2.copy_result(value: object) -> object
EvalResultReportV2.validate_report() -> EvalResultReportV2

Release source · Runtime guarantees · Capability guides

EvalStructuredJudgeCriterionPresentationV1

Section titled “EvalStructuredJudgeCriterionPresentationV1”
from cayu.evals import EvalStructuredJudgeCriterionPresentationV1

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
criterion_id StrictStr required
weight StrictStr required
score StrictStr required
weighted_contribution StrictStr required
explanation StrictStr | None required
explanation_state Literal['available', 'redacted', 'unavailable'] required
EvalStructuredJudgeCriterionPresentationV1.validate_criterion_id(value: str, info) -> str
EvalStructuredJudgeCriterionPresentationV1.validate_decimal(value: str, info) -> str

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalStructuredJudgePresentationV1

Decision-useful safe view of one admitted structured judgment.

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
detail PublishedStructuredModelJudgeDetail required
threshold_passed StrictBool | None required
criteria tuple[EvalStructuredJudgeCriterionPresentationV1, ...] Field(default=(), max_length=EVAL_CORPUS_MAX_JUDGE_CRITERIA)
EvalStructuredJudgePresentationV1.copy_detail(value: object) -> object
EvalStructuredJudgePresentationV1.validate_criteria_order(value: object, info) -> object
EvalStructuredJudgePresentationV1.validate_criteria_match_detail(value: tuple[EvalStructuredJudgeCriterionPresentationV1, ...], info) -> tuple[EvalStructuredJudgeCriterionPresentationV1, ...]
EvalStructuredJudgePresentationV1.validate_threshold_passed(value: bool | None, info) -> bool | None

Release source · Runtime guarantees · Capability guides

from cayu.evals import EvalTrialPresentationV1

Bases: _PortableModel.

Defaults below are source expressions; Field(...) includes schema/validation settings and default_factory creates a fresh value. Inherited declarations are included. Constructor behavior follows the model or dataclass shown in the bases/decorators.

Name Type Default
execution_failure_category Literal['provider_failure', 'environment_failure', 'execution_failure'] | None Field(default=None, exclude_if=lambda value: value is None)
execution_status Literal['completed', 'failed'] | None Field(default=None, exclude_if=lambda value: value is None)
trial_number StrictInt | None Field(default=None, ge=1, le=EVAL_CORPUS_MAX_TRIALS)
status PublishedStatus required
score StrictFloat | None Field(default=None, ge=0.0, le=1.0)
diagnostic_code EvalTrialDiagnosticCode | None None
dimensions EvalResultOutcomeDimensionsV1 required
assertions tuple[EvalAssertionPresentationV1, ...] Field(min_length=1, max_length=EVAL_CORPUS_MAX_ASSERTIONS_PER_CASE)
EvalTrialPresentationV1.validate_assertions_order(value: object, info) -> object
EvalTrialPresentationV1.validate_trial() -> EvalTrialPresentationV1

Release source · Runtime guarantees · Capability guides

from cayu.evals import present_eval_result

Compile one immutable public result into its canonical explainable view.

present_eval_result(result: CorpusExecutionResult | CapturedEvaluationResultV1) -> EvalResultPresentationV2

Release source · Runtime guarantees · Capability guides