PyHHMM: A Python Library for Heterogeneous Hidden Markov Models

Abstract

We introduce PyHHMM, an object-oriented open-source Python implementation of Heterogeneous-Hidden Markov Models (HHMMs). In addition to HMM's basic core functionalities, such as different initialization algorithms and classical observations models, i.e., continuous and multinoulli, PyHHMM distinctively emphasizes features not supported in similar available frameworks: a heterogeneous observation model, missing data inference, different model order selection criterias, and semi-supervised training. These characteristics result in a feature-rich implementation for researchers working with sequential data. PyHHMM relies on the numpy, scipy, scikit-learn, and seaborn Python packages, and is distributed under the Apache-2.0 License. PyHHMM's source code is publicly available on Github 1 to facilitate adoptions and future contributions. A detailed documentation 2 , which covers examples of use and models' theoretical explanation, is available. The package can be installed through the Python Package Index (PyPI).

Single-Cell DNA Methylation Analysis of Chicken Lampbrush Chromosomes

“…To define train and test subsets, we selected 54 non-overlapping loci, 300 kb length each, with randomly chosen start positions so that all three types of chicken chromosomes (macro-chromosomes, inter-chromosomes and micro-chromosomes) had the same representation in train and test sets. The AIC and BIC metrics from HMM implementation [ 25 , 28 ] were used to find the optimal parameters for MethylationHMM (n_states = 3, distance_param = 1250). These optimal parameters were used to make further decoding of oocyte methylation states.…”

Sleep Activity Recognition and Characterization from Multi-Source Passively Sensed Data

“…To manage the heterogeneity of our input sources, we developed a variation of ordinary HMMs which we refer to as Heterogeneous Hidden Markov Model (HHMM) 1 [48]. The proposed HHMM, which could be employed for any HAR task [49,42], can simultaneously manage categorical and continuous data thanks to its observation model, which is visible in Fig.…”