autointent.metrics.retrieval_precision#

autointent.metrics.retrieval_precision(query_labels, candidates_labels, k=None)#

Calculate the precision at position k.

Precision at position $k$ is calculated as:

$$\text{Precision@k}=\frac{1}{N}\sum _{i=1}^N\frac{|y_{\text{query},i}\cap y_{\text{candidates},i}^{(1:k)}|}{k}$$

where:

• $N$ is the total number of queries,

• $y_{\text{query},i}$ is the true label for the $i$-th query,

• $y_{\text{candidates},i}^{(1:k)}$ is the set of top-k predicted labels for the $i$-th query.

Parameters:

• query_labels (autointent.metrics.custom_types.LABELS_VALUE_TYPE) – For each query, this list contains its class labels

• candidates_labels (autointent.metrics.custom_types.CANDIDATE_TYPE) – For each query, these lists contain class labels of items ranked by a retrieval model (from most to least relevant)

• k (int | None) – Number of top items to consider for each query

Returns:

Score of the retrieval metric

Return type:

float