recbole.evaluator.proxy_evaluator¶

class recbole.evaluator.proxy_evaluator.ProxyEvaluator(config)[source]¶

Bases: object

ProxyEvaluator is used to assign the corresponding evaluator according to the evaluation metrics, for example, TopkEvaluator for top-k metrics, and summarize the results of all evaluators.

build()[source]¶

assign evaluators according to metrics.

Returns: a list of evaluators.
Return type: list

collect(interaction, scores)[source]¶

collect the all used evaluators’ intermediate result of one batch.

Parameters

interaction (Interaction) – AbstractEvaluator of the batch
scores (tensor) – the tensor of model output with size of (N, )

evaluate(batch_matrix_list, eval_data)[source]¶

calculate the metrics of all batches. It is called at the end of each epoch

Parameters

batch_matrix_list (list) – the results of all batches
eval_data (Dataset) – the class of test data

Returns

such as {'Hit@20': 0.3824, 'Recall@20': 0.0527, 'Hit@10': 0.3153, 'GAUC': 0.9236}

Return type

dict

merge_batch_result(batch_matrix_list)[source]¶

merge all the intermediate result got in self.collect for used evaluators separately.

Parameters: batch_matrix_list (list) – the results of all batches not separated
Returns: used evaluators’ results of all batches
Return type: dict