> This page is for version Main (default). > For other versions, use one of these documentation indexes: > - Main (default): https://docs.nvidia.com/nemo/gym/main/llms.txt > - 0.7.0: https://docs.nvidia.com/nemo/gym/v0.7.0/llms.txt > - 0.6.0: https://docs.nvidia.com/nemo/gym/v0.6.0/llms.txt > - 0.5.1: https://docs.nvidia.com/nemo/gym/v0.5.1/llms.txt > - 0.5.0: https://docs.nvidia.com/nemo/gym/v0.5.0/llms.txt > - 0.4.0: https://docs.nvidia.com/nemo/gym/v0.4.0/llms.txt > - 0.3.0: https://docs.nvidia.com/nemo/gym/v0.3.0/llms.txt > - 0.2.1: https://docs.nvidia.com/nemo/gym/v0.2.1/llms.txt > For clean Markdown of any page, append .md to the page URL. > For a complete documentation index, see https://docs.nvidia.com/nemo/gym/llms.txt. > For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.nvidia.com/nemo/gym/_mcp/server. # nemo_gym.rollouts.health.types Data contracts shared by rollout-health checks and orchestration. ## Module Contents ### Classes | Name | Description | | ------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------- | | [`CheckInput`](#nemo_gym-rollouts-health-types-CheckInput) | Persisted or derived evidence required to evaluate a check. | | [`CheckScope`](#nemo_gym-rollouts-health-types-CheckScope) | - | | [`CheckSpec`](#nemo_gym-rollouts-health-types-CheckSpec) | Stable, self-describing health-check contract. | | [`CheckSubject`](#nemo_gym-rollouts-health-types-CheckSubject) | - | | [`Finding`](#nemo_gym-rollouts-health-types-Finding) | Evidence emitted by a check. Checks never emit verdicts. | | [`HealthCheckResult`](#nemo_gym-rollouts-health-types-HealthCheckResult) | - | | [`RolloutDigest`](#nemo_gym-rollouts-health-types-RolloutDigest) | - | | [`RunHealthConfig`](#nemo_gym-rollouts-health-types-RunHealthConfig) | The health-check settings every command that judges a run shares: one place for names and defaults. | | [`RunHealthLimits`](#nemo_gym-rollouts-health-types-RunHealthLimits) | The run-level failure limits, as a share of expected rollouts; `None` is no limit, the default. | | [`_AgentStep`](#nemo_gym-rollouts-health-types-_AgentStep) | - | | [`_CallBindings`](#nemo_gym-rollouts-health-types-_CallBindings) | - | | [`_LineSlice`](#nemo_gym-rollouts-health-types-_LineSlice) | - | | [`_TaskRepeat`](#nemo_gym-rollouts-health-types-_TaskRepeat) | - | | [`_WorkerInput`](#nemo_gym-rollouts-health-types-_WorkerInput) | - | ### Data [`CALL_BINDING_INPUTS`](#nemo_gym-rollouts-health-types-CALL_BINDING_INPUTS) [`LimitsSource`](#nemo_gym-rollouts-health-types-LimitsSource) [`ROLLOUT_ID_KEY`](#nemo_gym-rollouts-health-types-ROLLOUT_ID_KEY) [`ROLLOUT_INDEX_KEY`](#nemo_gym-rollouts-health-types-ROLLOUT_INDEX_KEY) [`TASK_INDEX_KEY`](#nemo_gym-rollouts-health-types-TASK_INDEX_KEY) [`Verdict`](#nemo_gym-rollouts-health-types-Verdict) [`__getattr__`](#nemo_gym-rollouts-health-types-__getattr__) ### API ```python class nemo_gym.rollouts.rollout_health.CheckInput ``` **Bases:** `enum.Enum` Persisted or derived evidence required to evaluate a check. **`AGENT_TURNS`** `= 'agent_turns'` --- **`BOUND_CALLS`** `= 'bound_calls'` --- **`OBSERVED_MODEL_CALLS`** `= 'observed_model_calls'` --- **`OWNED_MODEL_CALLS`** `= 'owned_model_calls'` --- **`RECORD`** `= 'record'` --- **`REPEAT_DIGESTS`** `= 'repeat_digests'` --- **`REPEAT_VERDICTS`** `= 'repeat_verdicts'` --- **`RUN_ACCOUNTING`** `= 'run_accounting'` --- **`TRAJECTORY`** `= 'trajectory'` --- ```python class nemo_gym.rollouts.rollout_health.CheckScope ``` **Bases:** `enum.Enum` **`ROLLOUT`** `= 'rollout'` --- **`RUN`** `= 'run'` --- **`TASK`** `= 'task'` --- ```python class nemo_gym.rollouts.rollout_health.CheckSpec() ``` **Bases:** `BaseModel` Stable, self-describing health-check contract. **`evaluation_scope`** `CheckScope` --- **`id`** `str` --- **`model_config`** `= ConfigDict(frozen=True)` --- **`reads`** `frozenset[CheckInput]` --- **`subject`** `CheckSubject` --- ```python nemo_gym.rollouts.rollout_health.CheckSpec._require_one_call_binding_view( reads: frozenset[nemo_gym.rollouts.health.types.CheckInput] ) -> frozenset[nemo_gym.rollouts.health.types.CheckInput] ``` classmethod ```python class nemo_gym.rollouts.rollout_health.CheckSubject ``` **Bases:** `enum.Enum` **`AGENT_TURN`** `= 'agent_turn'` --- **`CHECK_EXECUTION`** `= 'check_execution'` --- **`MODEL_CALL`** `= 'model_call'` --- **`RECORD`** `= 'record'` --- **`ROLLOUT`** `= 'rollout'` --- **`RUN`** `= 'run'` --- **`TASK`** `= 'task'` --- **`TRAJECTORY_CAPTURE`** `= 'trajectory_capture'` --- ```python class nemo_gym.rollouts.rollout_health.Finding() ``` **Bases:** `BaseModel` Evidence emitted by a check. Checks never emit verdicts. **`check`** `str` --- **`detail`** `dict[str, Any] = Field(default_factory=dict)` --- **`locator`** `dict[str, int | str] | None = None` --- **`subject`** `dict[str, int | str]` --- ```python class nemo_gym.rollouts.rollout_health.HealthCheckResult() ``` **Bases:** `BaseModel` **`rollouts`** `list[RolloutDigest]` --- **`summary`** `dict[str, Any]` --- **`summary_path`** `Path` --- **`verdicts_path`** `Path` --- ```python class nemo_gym.rollouts.rollout_health.RolloutDigest() ``` **Bases:** `BaseModel` **`capture_completion_tokens`** `int = 0` --- **`capture_observed`** `bool` --- **`capture_prompt_tokens`** `int = 0` --- **`context_window_exceeded`** `bool | None = None` --- **`duplicated_calls`** `int = 0` --- **`ended_on_error`** `bool = False` --- **`errors_by_status`** `dict[str, int] = Field(default_factory=dict)` --- **`findings`** `list[Finding]` --- **`model_call_errors`** `int = 0` --- **`model_call_errors_usage_unknown`** `int = 0` --- **`model_calls`** `int = 0` --- **`policy_calls_observed`** `bool = False` --- **`responses`** `dict[str, list[str]] | None = None` --- **`rollout_id`** `str` --- **`rollout_index`** `int | str` --- **`scored`** `bool = False` --- **`stage_index`** `int | str | None = None` --- **`successful_model_calls`** `int = 0` --- **`task_index`** `int | str` --- **`timing`** `RolloutTiming | None = None` --- **`transcript_completion_tokens`** `int = 0` --- **`transcript_prompt_tokens`** `int = 0` --- **`unobserved`** `list[str]` --- **`upstream_recovery_observations`** `dict[str, int] | None = None` --- **`verdict`** `Verdict` --- **`verification_skipped`** `bool | None = None` --- ```python class nemo_gym.rollouts.health.types.RunHealthConfig() ``` **Bases:** `BaseModel` The health-check settings every command that judges a run shares: one place for names and defaults. **`health_check_max_final_failure_fraction`** `float | None` --- **`health_check_max_not_final_failure_fraction`** `float | None` --- **`run_health_limits`** `RunHealthLimits` --- ```python class nemo_gym.rollouts.health.types.RunHealthLimits() ``` **Bases:** `BaseModel` The run-level failure limits, as a share of expected rollouts; `None` is no limit, the default. **`max_final_failure_fraction`** `float | None = Field(default=None, ge=0, le=1)` --- **`max_not_final_failure_fraction`** `float | None = Field(default=None, ge=0, le=1)` --- **`model_config`** `= ConfigDict(frozen=True)` --- ```python nemo_gym.rollouts.health.types.RunHealthLimits.summary( source: nemo_gym.rollouts.health.types.LimitsSource | collections.abc.Mapping[str, nemo_gym.rollouts.health.types.LimitsSource] ) -> dict[str, typing.Any] ``` The limits as written to `quality_summary.json`, so the verdict can be reproduced. `source` maps each limit to where it came from; a single source applies to every limit. ```python class nemo_gym.rollouts.health.types._AgentStep( locator: dict[str, int | str], has_message: bool, has_tool_calls: bool, model_call_refs: tuple[str, ...] ) ``` Dataclass **`has_message`** `bool` --- **`has_model_activity`** `bool` --- **`has_tool_calls`** `bool` --- **`locator`** `dict[str, int | str]` --- **`model_call_refs`** `tuple[str, ...]` --- ```python class nemo_gym.rollouts.health.types._CallBindings( references: tuple[str, ...], matched_calls: tuple[dict[str, typing.Any], ...], missing_references: tuple[str, ...], duplicated_references: tuple[tuple[str, int], ...], resolutions: collections.abc.Mapping[tuple[str, str, str, str], tuple[dict[str, typing.Any], ...]] = dict() ) ``` Dataclass **`complete`** `bool` --- **`duplicated_references`** `tuple[tuple[str, int], ...]` --- **`matched_calls`** `tuple[dict[str, Any], ...]` --- **`missing_references`** `tuple[str, ...]` --- **`observed`** `bool` --- **`references`** `tuple[str, ...]` --- **`resolutions`** `Mapping[tuple[str, str, str, str], tuple[dict[str, Any], ...]] = field(default_factory=dict)` --- ```python class nemo_gym.rollouts.health.types._LineSlice( path: str, offset: int, length: int, ordinal: int, source_index: int, line_number: int, file_identity: tuple[int, int], expected_identity: tuple[str | None, str, int] | None = None, legacy_attempt_index: int | None = None ) ``` Dataclass **`expected_identity`** `tuple[str | None, str, int] | None = None` --- **`file_identity`** `tuple[int, int]` --- **`legacy_attempt_index`** `int | None = None` --- **`length`** `int` --- **`line_number`** `int` --- **`offset`** `int` --- **`ordinal`** `int` --- **`path`** `str` --- **`source_index`** `int` --- ```python class nemo_gym.rollouts.health.types._TaskRepeat( rollout_index: int | str, verdict: nemo_gym.rollouts.health.types.Verdict, policy_calls_observed: bool, successful_model_calls: int ) ``` Dataclass **`policy_calls_observed`** `bool` --- **`rollout_index`** `int | str` --- **`successful_model_calls`** `int` --- **`verdict`** `Verdict` --- ```python class nemo_gym.rollouts.health.types._WorkerInput( line: nemo_gym.rollouts.health.types._LineSlice, ignored_checks: frozenset[str] ) ``` Dataclass **`ignored_checks`** `frozenset[str]` --- **`line`** `_LineSlice` --- ```python nemo_gym.rollouts.health.types.CALL_BINDING_INPUTS = frozenset({CheckInput.BOUND_CALLS, CheckInput.OWNED_MODEL_CALLS}) ``` ```python nemo_gym.rollouts.health.types.LimitsSource = Literal['config', 'persisted', 'override'] ``` ```python nemo_gym.rollouts.health.types.ROLLOUT_ID_KEY = ROLLOUT_ID_KEY_NAME ``` ```python nemo_gym.rollouts.health.types.ROLLOUT_INDEX_KEY = ROLLOUT_INDEX_KEY_NAME ``` ```python nemo_gym.rollouts.health.types.TASK_INDEX_KEY = TASK_INDEX_KEY_NAME ``` ```python nemo_gym.rollouts.health.types.Verdict = Literal['healthy', 'unhealthy', 'unobserved'] ``` ```python nemo_gym.rollouts.health.types.__getattr__ = moved_names_getattr(__name__) ```