TODO#
Todo
Support for advanced rating situations is not yet supported:
repeated ratings
mixed implicit & explicit feedback
later actions removing earlier ratings
(The original entry is located in /home/runner/work/lkpy/lkpy/lenskit/lenskit/data/dataset.py:docstring of lenskit.data.dataset.Dataset, line 8.)
Todo
Support for item and user content or metadata is not yet implemented.
(The original entry is located in /home/runner/work/lkpy/lkpy/lenskit/lenskit/data/dataset.py:docstring of lenskit.data.dataset.Dataset, line 15.)
Todo
Aggregate is currently ignored because repeated interactions are not yet supported.
(The original entry is located in /home/runner/work/lkpy/lkpy/lenskit/lenskit/data/dataset.py:docstring of lenskit.data.dataset.Dataset.interaction_matrix, line 6.)
Todo
Right now, selection / subsetting only happens on the CPU, and will move data to the CPU for the subsetting operation. There is no reason, in principle, why we cannot subset on GPU. Future revisions may add support for this.
(The original entry is located in /home/runner/work/lkpy/lkpy/lenskit/lenskit/data/items.py:docstring of lenskit.data.items.ItemList, line 45.)
Todo
Document and test methods for extending the recommendation query with arbitrary data to be used by client-provided pipeline components.
(The original entry is located in /home/runner/work/lkpy/lkpy/lenskit/lenskit/data/query.py:docstring of lenskit.data.query.RecQuery, line 5.)
Todo
When LensKit supports context-aware recommendation, this should be extended to include context cues.
(The original entry is located in /home/runner/work/lkpy/lkpy/lenskit/lenskit/data/query.py:docstring of lenskit.data.query.RecQuery, line 9.)
Todo
Repeated interactions are not yet supported.
(The original entry is located in /home/runner/work/lkpy/lkpy/lenskit/lenskit/data/convert.py:docstring of lenskit.data.convert.from_interactions_df, line 4.)
Todo
Right now, this uses the ‘rating’ as a count. Actually use counts.
(The original entry is located in /home/runner/work/lkpy/lkpy/lenskit-hpf/lenskit/hpf.py:docstring of lenskit.hpf.HPFScorer, line 4.)
Todo
This chapter needs to be rewritten for 2025.1 (in progress).
(The original entry is located in /home/runner/work/lkpy/lkpy/docs/guide/documenting.rst, line 4.)
Todo
Add example (and simplified ergonomics) for fallback prediction.
(The original entry is located in /home/runner/work/lkpy/lkpy/docs/guide/evaluation/predictions.rst, line 73.)
Todo
Write more tutorial / user manual documentation for the item lists.
(The original entry is located in /home/runner/work/lkpy/lkpy/docs/guide/item-lists.rst, line 15.)
Todo
Update for 2025.1.
(The original entry is located in /home/runner/work/lkpy/lkpy/docs/guide/performance.rst, line 7.)
Todo
Redo some of those types with user & item data, etc.
(The original entry is located in /home/runner/work/lkpy/lkpy/docs/guide/pipeline.rst, line 26.)
Todo
Provide utility functions to make more common wiring operations easy so there is middle ground between “give me a standard pipeline” and “make me do everything myself”.
(The original entry is located in /home/runner/work/lkpy/lkpy/docs/guide/pipeline.rst, line 29.)
Todo
Rethink the “keyword inputs only” constraint in view of the limitation it places on fallback or other compositional components — it’s hard to specify a component that implements fallback logic for an arbitrary number of inputs.
(The original entry is located in /home/runner/work/lkpy/lkpy/docs/guide/pipeline.rst, line 34.)
Todo
Document queries.
(The original entry is located in /home/runner/work/lkpy/lkpy/docs/guide/queries.rst, line 8.)