calc_confusions
- rectools.metrics.classification.calc_confusions(merged: DataFrame, k: int) DataFrame [source]
Calculate some intermediate metrics from prepared data (it’s a helper function).
- For each user (Columns.User) the following metrics are calculated:
LIKED - number of items the user has interacted (bought, liked) with;
TP - number of relevant recommendations among the first k items at the top of recommendation list;
FP - number of non-relevant recommendations among the first k items of recommendation list;
FN - number of items the user has interacted with but that weren’t recommended (in top k).
- Parameters
merged (pd.DataFrame) – Result of merging recommendations and interactions tables. Can be obtained using merge_reco function.
k (int) – Number of items at the top of recommendations list that will be used to calculate metric.
- Returns
Table with columns: Columns.User, LIKED, TP, FP, FN.
- Return type
pd.DataFrame
Notes
left = all - K TP = sum(rank) FP = K - TP FN = liked - TP TN = all - K - FN = left - FN = left - liked + TP