1-D step detection algorithms
Project description
scikit-step
1-D step detection algorithms.
Example
Basic usage
from skstep import GaussStepFinder
sf = GaussStepFinder() # initialize
result = sf.fit(data) # fitting
result.plot() # plot result
result.step_positions # step positions
result.means # mean values at each constant region
result.step_sizes # change between adjacent constant regions
result.lengths # length of each constant region
Chunkwise fitting
Computation time of step finding algorithms are usually around O(N^1.5). This means that fragmenting large data makes computation faster while does not affect the result a lot.
All the step finding algorithms are implemented with chunkwise fitting with parallel processing using dask.
sf.fit_chunkwise(data)
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
scikit_step-0.1.0.tar.gz
(9.2 kB
view hashes)
Built Distribution
Close
Hashes for scikit_step-0.1.0-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 3d68fe799b80e5c24d98a8a89a289ca87fd912a75f51cb58b442ba597b4cedf0 |
|
MD5 | 6a0119a9dc5b3a6eb2332fd0b2e5126f |
|
BLAKE2b-256 | 235548b14165d17e8af7217bff3c7d157d086e8f2637bd7c188eefc17775c149 |