This folder contains the code used to run analyses and generate figures in the paper "The Gaussian-Linear Hidden Markov model: a Python package". doi:
https://doi.org/10.48550/arXiv.2312.07151
Files in this repository:
- MEG.py : a python analysis script loading and running the GLHMM analyses on a MEG dataset involving human subjects performing a visual-memory task (see paper for details on the analyses and data). The script assumes data are pre-downloaded. It outputs analyses results as .npy files.
- ecog.py : a python analysis script loading and running the GLHMM analyses on an EGog dataset involving monkeys performing a motor task (see paper for details on the analyses and data). The script assumes data are pre-downloaded. It outputs analyses results as .npy files.
- FIG2_ecog_plot.py : a python script loading the results of the analyses on ecog data and plotting them. The script assumes the analysis results are correctly stored. It outputs plots in paper figure 2.
- GLHMM_fig1.py : a python script running the GLHMM and a standard HMM LFP and spike data of rats performing an odour-memory task. The script assumes the data are downloaded and preprocessed (preprocessing script not included, but available as part of a different project). It outputs plots in paper FIGURE 1.
- FIG4_MEG_plot.py : a python script loading the results of the analyses on MEG data and plotting them. The script assumes the analysis results are correctly stored. It outputs plots in paper figure 4.
- prediction_tutorial.ipynb : a jupyter notebook script containing a tutorial on how to use the GLHMM toolbox. It also runs analyses on the HCP dataset (human fMRI, assuming data are pre-downloaded) and plots its results, as in paper figure 6.
- run_glhmm.sh : a shell script running and measuring the computational performance of stochastic and non-stochastic GLHMM training on MEG and fMRI data.
- Stoc_v_NonStoc.py : a python script called by "run_glhmm.sh", performing stochastic and non stochastic training of GLHMM on MEG and fMRI data.
- plotting_Fig7.py : a python script loading the results of stochastic and non-stochastic inference and plotting them. It outputs plots for FIGURE 7 in the main paper and Supplementary Figures S1 and S2.
- plotting_suppl.py : a python script loading the results of stochastic and non-stochastic inference and plotting them. It outputs plots for Supplementary Figures S3 and S4.
Please note: the datasets analyzed are not provided within this repository. In order to be able to run the analyses, please make sure you have access to the required data.
Disclaimer: this is not the toolbox repository, neither the official documentation website for the toolbox!
Find the toolbox at: https://github.com/vidaurre/glhmm/glhmm
Tutorials and examples at: https://github.com/vidaurre/glhmm/tree/main/docs/notebooks
Toolbox documentation at: https://glhmm.readthedocs.io/en/latest/index.html
If you have questions, issues, or suggestions, feel free to reach us at laurama@cfin.au.dk or use the Discussions section in the toolbox github repo: https://github.com/vidaurre/glhmm