Implementation of `Precision` Metric. Basic RecSys metric that calculates precision of `Top-K` elements. Parameters ---------- t

Implementation of `Recall` Metric. Basic RecSys metric that calculates recall of `Top-K` elements. Parameters ---------- topk :

Implementation of `Normalized Discounter Cumulative Gain` Metric. Graded relevance as a measure of usefulness, or gain, from examining a set

Implementation of Adaptive Negative Sampling from `Improving Pairwise Learning for Item Recommendation from Implicit Feedback`. Paramete

Collate Function as an object. Parameters ---------- pad : `list[str] | None`, optional (default = None) List of keys to app

A Click type to represent dictionary as parameters for command.

The BPR pairwise loss. Parameters ---------- size_average : `bool`, optional (default = True) By default, the losses are ave

Implementation of `F-Beta` Metric. Basic RecSys metric that calculates precision of `Top-K` elements. Parameters ---------- topk

In Memory dataset from JsonLines. Parameters ---------- path : `Path | str`, required File path.

Iterable dataset from JsonLines. Parameters ---------- path : `Path | str`, required File path.

Implementation of `Mean Average Precision` Metric. The precision metric summarizes the fraction of relevant items out of the whole the re

Stacked FeedForward Layers. Parameters ---------- linears : `list[torch.nn.Linear]`, required A collection of linear transfo

Interface for metrics with masked computation.

Base class for all metrics.

The BPR model. Parameters ---------- logits_model : `BaseLogitModel`, required The model that produces logits for a user-ite

The Global Item-Popularity model. Parameters ---------- num_items : `int`, required Number of items in the dataset.

Implementation of Uniform Negative Sampling. Parameters ---------- num_items : `int`, required Number of items in the datase