Clustered Learning of Approximate Manifolds
Project description
CLAM: Clustered Learning of Approximate Manifolds (v0.12.0)
CLAM is a Rust/Python library for learning approximate manifolds from data. It is designed to be fast, memory-efficient, easy to use, and scalable for big data applications.
CLAM provides utilities for fast search (CAKES) and anomaly detection (CHAODA).
As of writing this document, the project is still in a pre-1.0 state. This means that the API is not yet stable and breaking changes may occur frequently.
Installation
> python3 -m pip install "abd_clam==0.12.0"
Usage
from abd_clam.search import CAKES
from abd_clam.utils import synthetic_data
# Get the data.
data, _ = synthetic_data.bullseye()
# data is a numpy.ndarray in this case but it could just as easily be a numpy.memmap if your data do fit in RAM.
# We used numpy memmaps for the research, though they impose file-IO costs.
model = CAKES(data, 'euclidean')
# The Search class provides the functionality described in our [CHESS paper](https://arxiv.org/abs/1908.08551).
model.build(max_depth=50)
# Build the search tree to depth of 50.
# This method can be called again with a higher depth, if needed.
query, radius = data[0], 0.5
rnn_results = model.rnn_search(query, radius)
# This is how we perform ranged nearest neighbors search with radius 0.5 around the query.
# The results are returned as a dictionary whose keys are indices into the data array and whose values are the distance to the query.
knn_results = model.knn_search(query, 10)
# This is how we perform k-nearest neighbors search for the 10 nearest neighbors of query.
# TODO: Provide snippets for using CHAODA
License
Citation
TODO
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
abd_clam-0.12.0.tar.gz
(34.7 kB
view hashes)
Built Distribution
abd_clam-0.12.0-py3-none-any.whl
(42.8 kB
view hashes)
Close
Hashes for abd_clam-0.12.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 5369e550a76e842feb6f77a634e3e0e0ac968f24c0d65d8df6330e7c9f958496 |
|
MD5 | b11992d4af4bfc9cb7c7bee7593d5160 |
|
BLAKE2b-256 | 5e47505268320291bf247269b1223e466489c7fea2da33692b7ee848e6dc58a5 |