=== FAQ === Does LightFM have a GPU-based implementation? ========== No, there is no option to run training or inference on the GPU with LightFM. There are currently no plans to change this. See ht...

Welcome to LightFM's documentation! ========== LightFM is a Python implementation of a number of popular recommendation algorithms for both implicit and explicit feedback. It also makes it possible...

Datasets ========== .. autofunction:: lightfm.datasets.movielens.fetch_movielens .. autofunction:: lightfm.datasets.stackexchange.fetch_stackexchange

An implicit feedback recommender for the Movielens dataset ========== Implicit feedback ---------- For some time, the recommender system literature focused on explicit feedback: the Netflix prize f...

Using different learning schedules ========== ``lightfm`` implements two learning schedules: adagrad and adadelta. Neither is clearly superior, and, like other hyperparameter choices, the best lear...

Model evaluation ========== .. automodule:: lightfm.evaluation :members: :undoc-members: :show-inheritance:

Item cold-start: recommending StackExchange questions ========== In this example we'll use the StackExchange dataset to explore recommendations under item-cold start. Data dumps from the StackExcha...

Quickstart ========== In this example, we'll build an implicit feedback recommender using the Movielens 100k dataset (http://grouplens.org/datasets/movielens/100k/). The code behind this example is...

Learning-to-rank using the WARP loss ========== LightFM is probably the only recommender package implementing the WARP (Weighted Approximate-Rank Pairwise) loss for implicit feedback learning-to-ra...

Building datasets ========== In this example, we'll use LightFM's built-in ``Dataset`` class to build an interaction dataset from raw data. The goal is to demonstrate how to go from raw data (lists...