scikitplot.annoy

High-level Python interface for the C++ Annoy backend.

Spotify Annoy (Approximate Nearest Neighbors Oh Yeah).

Exports:

• Annoy → low-level C-extension type (stable)

• AnnoyBase → alias of annoylib.Annoy Index

• Index → high-level Python subclass (picklable)

• AnnoyIndex → public aliases of Annoy Index

See also

• ANNoy (experimental)

• spotify/annoy

• https://pypi.org/project/annoy

Note

• https://www.sandgarden.com/learn/faiss

• https://www.sandgarden.com/learn/annoy-approximate-nearest-neighbors-oh-yeah

See also

• https://en.wikipedia.org/wiki/Nearest_neighbor_search

• https://www.researchgate.net/publication/386374637_Optimizing_Domain-Specific_Image_Retrieval_A_Benchmark_of_FAISS_and_Annoy_with_Fine-Tuned_Features

• https://www.researchgate.net/publication/363234433_Analysis_of_Image_Similarity_Using_CNN_and_ANNOY

• https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/XboxInnerProduct.pdf

• https://link.springer.com/chapter/10.1007/978-981-97-7831-7_2

References

[1]

Spotify AB. (2013, Feb 20). “Approximate Nearest Neighbors Oh Yeah” Github. https://pypi.org/project/annoy

Examples

>>> # from annoy import AnnoyIndex
>>> from scikitplot.annoy import Annoy, AnnoyBase, AnnoyIndex, Index
>>> f = 40  # vector dimensionality
>>> t = AnnoyIndex(f, "angular")  # Length of item vector and metric
>>> t.add_item(0, [1] * f)
>>> t.build(10)  # Build 10 trees
>>> t.get_nns_by_item(0, 1)  # Find nearest neighbor

User guide. See the ANNoy section for further details.

ANNoy (Approximate Nearest Neighbors Oh Yeah)

Annoy	Compiled with GCC/Clang.
AnnoyBase	alias of Annoy
AnnoyIndex	alias of Index
Index	High-level Pythonic Annoy wrapper with picklable (or pickle-able).

ANNoy Mixins

ManifestMixin	Export/import metadata only.
ObjectIOMixin	Persistence for the Python object (Index/Annoy alias), not the raw Annoy index file.
PathAwareAnnoy	Thin Python subclass that tracks the last known on-disk path.
PickleMixin	Adds strict persistence support.