https://github.com/hauselin/megmvpa

MEG MVPA Tutorial

https://github.com/hauselin/megmvpa

Repository

MEG MVPA Tutorial

https://github.com/hauselin/megmvpa/blob/master/

# MEG MVPA Tutorial (Python & Matlab)

This tutorial accompanies the paper titled "Multivariate pattern analysis for MEG: a comparison of dissimilarity measures", which is available [here](https://doi.org/10.1016/j.neuroimage.2018.02.044) ([preprint](https://doi.org/10.1101/172619)).

**Citation:** Guggenmos, M., Sterzer, P., & Cichy, R. M. (2018). Multivariate pattern analysis for MEG: a comparison of dissimilarity measures. NeuroImage. DOI: 10.1016/j.neuroimage.2018.02.044

## Python tutorial

### Preparation
This tutorial is based on [IPython/Jupyter Notebook](https://jupyter.org/) files, which are linked below. In addition, the tutorial can be downloaded as a [zip file](https://github.com/m-guggenmos/megmvpa/blob/master/tutorial_python.zip), which includes the notebook files, additional code files and the example dataset used for this tutorial. To reduce computational costs, the dataset is for one participant only and includes only 9 of 92 experimental conditions.

Content of the zip file:

File | Description
--- | --- 
cv.py | _containing code for pseudo-trials/permutations/cross-validation_
dissimilarity.py | _containing a number of custom dissimilarity measures_
weird.py | _weighted robust distance classifier (WeiRD), see also [here](https://github.com/m-guggenmos/weird)_
python_decoding.ipynb | Notebook on _Decoding_
python_reliability.ipynb | Notebook on _RDMs and Reliability_
python_distance.ipynb | Notebook on _Distance measures and cross-validation_
data01_sess1.npy | _data for subject 1, session 1_
data01_sess2.npy | _data for subject 1, session 2_
labels01_sess1.npy | _trial labels for subject 1, session 1_
labels01_sess2.npy | _trial labels for subject 1, session 2_

In addition, the tutorial requires 4 established scientific python packages: numpy, scipy, scikit-learn, matplotlib

### List of tutorials:
* [Decoding](https://github.com/m-guggenmos/megmvpa/blob/master/tutorial_python/python_decoding.ipynb)
* [RDMs and Reliability](https://github.com/m-guggenmos/megmvpa/blob/master/tutorial_python/python_reliability.ipynb)
* [Distance measures and cross-validation](https://github.com/m-guggenmos/megmvpa/blob/master/tutorial_python/python_distance.ipynb)

## Matlab tutorial

### Preparation
This tutorial is based on [IPython/Jupyter Notebook](https://jupyter.org/) files, which are linked below. In addition, the tutorial can be downloaded as a [zip file](https://github.com/m-guggenmos/megmvpa/blob/master/tutorial_matlab.zip), which includes the notebook files, additional code files and the example dataset used for this tutorial. To reduce computational costs, the dataset is for one participant only and includes only 9 of 92 experimental conditions.

Content of the zip file:

File | Description
--- | --- 
cov1para.m | _Shrinkage code (Ledoit & Wolf, 2004) for covariances_
weirdtrain.m & weirdpredict.m | _Weighted Robust Distance (WeiRD) classifier_
gnbtrain.m & gnbpredict.m | _Gaussian Naive Bayes (GNB) classifier_
matlab_decoding.ipynb | Notebook on _Decoding_
matlab_reliability.ipynb | Notebook on _RDMs and Reliability_
matlab_distance.ipynb | Notebook on _Distance measures and cross-validation_
data01_sess1.mat | _data for subject 1, session 1_
data01_sess2.mat | _data for subject 1, session 2_
labels01_sess1.mat | _trial labels for subject 1, session 1_
labels01_sess2.mat | _trial labels for subject 1, session 2_

In addition, the tutorial assumes a working [LIBSVM](https://www.csie.ntu.edu.tw/~cjlin/libsvm/#download) installation for Matlab.

### List of tutorials:
* [Decoding](https://github.com/m-guggenmos/megmvpa/blob/master/tutorial_matlab/matlab_decoding.ipynb)
* [RDMs and Reliability](https://github.com/m-guggenmos/megmvpa/blob/master/tutorial_matlab/matlab_reliability.ipynb)
* [Distance measures and cross-validation](https://github.com/m-guggenmos/megmvpa/blob/master/tutorial_matlab/matlab_distance.ipynb)

Owner

GitHub Events