A numerically-stable implementation of essential HMM algorithms with tutorials on training methods
Project description
Homomorph
The HMM class is fully documented. However, the tutorial is the recommended introduction to this package.
Dependencies
Homomorph is designed to be lightweight and only requires NumPy for array calculations and SciPy for its implementation of random variables. The minimum versions were set to the most recent at the time of initial release (1.17 and 1.8, respectively), but no functionality specific to these releases is needed to my knowledge.
Installation
To install Homomorph, run the following command:
pip install homomorph
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
homomorph-0.0.1.tar.gz
(6.7 kB
view hashes)
Built Distribution
Close
Hashes for homomorph-0.0.1-py3-none-any.whl
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | f1c503d9b793ea5f687c2dd9b8afb76f84cfde5e2f7206bfc8a2c5de5cd912d7 |
|
| MD5 | 6601aa4728ea8c250f9663a00d53f24f |
|
| BLAKE2b-256 | afaca9fe56b95f3699d29bc18dd8da8941ee7fd9b5e8685ebaf3be1bf8a3cd48 |
