Welcome to ssm’s documentation!
Contents
Welcome to ssm’s documentation!#
Important
We’re working full time on a JAX refactor of SSM that will take advantage of JIT compilation, GPU and TPU support, automatic differentation, etc. Please stay tuned for updates soon!
This package has fast and flexible code for simulating, learning, and performing inference in a variety of state space models. Currently, it supports:
Hidden Markov Models (HMM)
Auto-regressive HMMs (ARHMM)
Input-output HMMs (IOHMM)
Hidden Semi-Markov Models (HSMM)
Linear Dynamical Systems (LDS)
Switching Linear Dynamical Systems (SLDS)
Recurrent SLDS (rSLDS)
Hierarchical extensions of the above
Partial observations and missing data
We support the following observation models:
Gaussian
Student’s t
Bernoulli
Poisson
Categorical
Von Mises
HMM inference is done with either expectation maximization (EM) or stochastic gradient descent (SGD). For SLDS, we use stochastic variational inference (SVI).
- Hidden Markov Model Demo
- Simple Linear Dynamical System Demo
- Input Driven HMM
- Input Driven Observations (“GLM-HMM”)
- Switching Linear Dynamical System Demo
- Recurrent SLDS
- Poisson SLDS
- Poisson fLDS
- Variational Laplace-EM
- HMM State Clustering
- Cluster states from the original HMM
- Multi-population recurrent switching linear dynamical systems overview
- orphan
SSM Notebooks#
Notebooks to accompany the State Space Models (SSM) library.
_thumb.png)
