Stochastic Gradient MCMC for State Space Models

This repo contains the python code for stochastic gradient MCMC in state space models for the following papers: Stochastic Gradient MCMC for SSMs and "Stochastic Gradient MCMC for Nonlinear SSMs".

Overview

The sgmcmc_ssm folder contains the python module code.

The demo folder contains python scripts that demonstrate the sgmcmc_ssm module API. These scripts must be run with sgmcmc_ssm on the PYTHONPATH. For example, running ipython demo/<script name>.py from this project root folder.

Installation

Add the sgmcmc_ssm folder to the PYTHONPATH by (i) running code from the project root folder, (ii) adding the project root folder to the python path using sys.path.append(<path_to_sgmcmc_ssm_code>), or (iii) running setup.py.

Requirements: Python 3+, numpy, pandas, scipy, seaborn, joblib, scikit-learn,

Usage Examples

See demo/lgssm_quick_start.py for a quick overview of how to use the code. See demo/api_demos/ for additional details for each model.

Synthetic LGSSM Script Example

cd <path_to_sgmcmc_ssm_code>
python demo/synthetic/lgssm_demo.py

Ion-Channel Gaussian HMM Example

cd <path_to_sgmcmc_ssm_code>
python demo/ion_channel/ion_channel_subset_demo.py

Exchange Rate SVM Example

cd <path_to_sgmcmc_ssm_code>
python demo/exchange_rate/process_exchange_data.py
python demo/exchange_rate/exchange_rate_subset_demo.py

Meta

Christopher Aicher – aicherc [at] uw [dot] edu

Distributed under the MIT license. See LICENSE for more information.

https://github.com/aicherc/sgmcmc_ssm_code

Contributing

1. Fork it (https://github.com/aicherc/sgmcmc_ssm_code/fork)
2. Create your feature branch (git checkout -b feature/fooBar)
3. Commit your changes (git commit -am 'Add some fooBar')
4. Push to the branch (git push origin feature/fooBar)
5. Create a new Pull Request