config::loader::ConfigLoader¶

Public Functions¶

	Name
None	init(self self, Path project_root)
Dict[str, Any]	get_effective_config(self self, str niche)

Protected Functions¶

	Name
Dict[str, Any]	_load_global_config(self self)
Dict[str, Path]	_load_specialist_configs(self self)
None	_validate(self self)
None	_validate_endpoints(self self)
None	_validate_models(self self)
None	_validate_teacher(self self)
None	_validate_teacher_benchmark(self self)
None	_validate_pipeline_specialists(self self)
None	_validate_fp4_export(self self)
None	_apply_specialist_overrides(self self, Dict effective[str, Any], Dict specialist_data[str, Any])
Dict[str, Any]	_load_yaml(Path path)

Protected Attributes¶

	Name
	_project_root

Detailed Description¶

class config::loader::ConfigLoader;

Loads, validates, and resolves the two-layer pipeline configuration.

On construction, loads config/pipeline.yaml and all config/specialists/*.yaml
files. Validation runs immediately — a ConfigValidationError is raised for
any schema violation.

Public Functions Documentation¶

function init¶

None __init__(
    self self,
    Path project_root
)

function get_effective_config¶

Dict[str, Any] get_effective_config(
    self self,
    str niche
)

Return the effective config for *niche*, with per-specialist overrides applied.

The effective config starts as a deep copy of the global config. If a
specialist config exists for ``niche``, its values are deep-merged:
dict values merge recursively, lists and scalars replace the global
default.

Raises ConfigValidationError if *niche* is not in ``pipeline.specialists``.

Protected Functions Documentation¶

function _load_global_config¶

Dict[str, Any] _load_global_config(
    self self
)

function _load_specialist_configs¶

Dict[str, Path] _load_specialist_configs(
    self self
)

function _validate¶

None _validate(
    self self
)

function _validate_endpoints¶

None _validate_endpoints(
    self self
)

function _validate_models¶

None _validate_models(
    self self
)

function _validate_teacher¶

None _validate_teacher(
    self self
)

function _validate_teacher_benchmark¶

None _validate_teacher_benchmark(
    self self
)

function _validate_pipeline_specialists¶

None _validate_pipeline_specialists(
    self self
)

function _validate_fp4_export¶

None _validate_fp4_export(
    self self
)

Validate the fp4_export configuration block.

Per D-08: fp4_export is optional (Phase 1/2 may run without quantization).
When present, validates error_thresholds per block size, ternary_delta range,
min_block_size power-of-2, laplacian_levels, and log_mode_enabled type.

function _apply_specialist_overrides¶

None _apply_specialist_overrides(
    self self,
    Dict effective[str, Any],
    Dict specialist_data[str, Any]
)

Deep-merge specialist overrides into *effective* config in-place.```


### function _load_yaml

```python
static Dict[str, Any] _load_yaml(
    Path path
)

Protected Attributes Documentation¶

variable _project_root¶

_project_root;

Updated on 2026-07-25 at 22:56:57 +0000