Model Builder

model.build.build_evaluator(cfg) → Evaluator

Build an evaluator object based on the provided configuration.

Parameters:: cfg (dict) – The configuration dictionary containing the evaluator settings.
Returns:: An instance of the Evaluator class.
Return type:: Evaluator
Raises:: ValueError – If the metric specified in the configuration is not supported.

model.build.build_feature_extractor(cfg) → FeatureExtractor

Builds a feature extractor based on the configuration provided.

Parameters:: cfg (dict) – The configuration dictionary containing the model settings.
Returns:: The instantiated feature extractor object.
Return type:: FeatureExtractor
Raises:: ValueError – If the specified feature extractor type is not supported.

model.build.build_feature_predictor(cfg) → FeaturePredictor

Builds a feature predictor based on the configuration provided.

Parameters:: cfg (dict) – The configuration dictionary containing the model parameters.
Returns:: The built feature predictor object.
Return type:: FeaturePredictor
Raises:: ValueError – If the specified feature predictor type is not supported.

model.build.build_model(cfg) → ExtractorPredictor

Builds a model based on the provided configuration.

Parameters:: cfg (dict) – The configuration dictionary containing model parameters.
Returns:: The built model.
Return type:: model
Raises:: ValueError – If the model type specified in the configuration is not supported.

model.build.build_pretrained_model(cfg, weights_file) → ExtractorPredictor

Build a pretrained model based on the provided configuration and weights file.

Parameters:

Returns:

Pretrained model instance.

Return type:

ExtractorPredictor

Raises:

ValueError – If the specified model type is not supported.