pyhsmm
This Python library facilitates approximate unsupervised inference in Bayesian Hidden Markov Models (HMMs) and Hidden semi-Markov Models (HSMMs). It emphasizes Bayesian Nonparametric extensions, particularly the HDP-HMM and HDP-HSMM, utilizing weak-limit approximations. While innovative, users should note that this package is no longer maintained.
Top pyhsmm Alternatives
fungp
Fungp is an innovative genetic programming library designed for Clojure, facilitating the evolution of computer programs through processes inspired by biological evolution.
Statistiker
Statistiker offers a streamlined approach to statistics in Clojure, implementing widely-used algorithms for dataset analysis.
Vowpal Wabbit
Vowpal Wabbit (VW) is an advanced, fast out-of-core learning system developed through collaboration with Microsoft Research and Yahoo! Research.
Pebl
Pebl is a Python library and command line tool designed for learning the structure of Bayesian networks from prior knowledge and observations.
MachineLearning.jl
The MachineLearning.jl package is an innovative library crafted in pure Julia, designed to unify common machine learning algorithms with a consistent API.
CloudForest
This advanced machine learning software leverages ensembles of decision trees in pure Go, designed for tasks such as classification, regression, and feature selection.
Saul
It streamlines the creation of machine learning models, allowing programmers to focus on application logic...
Figaro
It offers built-in reasoning algorithms for drawing insights from data, seamlessly integrating with Scala and...
Swift AI
It boasts a suite of tools for artificial intelligence and scientific applications, complete with example...
warpt-ctc
Its fast parallel implementation on CPU and GPU significantly enhances scalability, making it ideal for...
htm.java
Designed for JavaSE 8 and above, it mirrors NuPIC's functionality while adhering to Java best...
Learning Based Java
It offers intuitive algorithms and robust frameworks tailored for data analysis and predictive modeling, enabling...
Naive Bayesian Classification for Golang
Featuring support for term frequency-inverse document frequency (TF-IDF) calculations, this library facilitates accurate classifications while...
Sparkling Water
It enables users to deploy custom AI models, monitor performance, and ensure compliance through a...
Swift Brain
It features advanced machine learning algorithms, including Bayes theorem and neural networks, alongside tools for...
Top pyhsmm Features
- Approximate unsupervised inference
- Bayesian Nonparametric extensions
- HDP-HMM integration
- Weak-limit approximations
- Supports explicit-duration HSMMs
- Gibbs sampling resampling loop
- Dynamic state inference
- Multiple observation sequences
- Custom observation distributions
- Memory-efficient model sampling
- Integration with Cython
- Compatibility with Python 3.7
- WeakLimitHDPHSMM instance creation
- Hyperparameter tuning options
- Visualization of posterior samples
- Supports custom duration distributions
- Comprehensive example directory.
