Loading Data into Unfold

Unfold is generally agnostic to how you load your data. You only require a Matrix (channel x time) or 3D-Array(channel x time x epochs) and an event-dataframe.

Setup

using Unfold
using UnfoldMakie,CairoMakie
using PyMNE
using DataFrames

MNE Demo Dataset

The easiest way to showcase this is to simply use a demo-dataset from MNE.

limo_epochs = PyMNE.datasets.limo.load_data(subject=1,path="~/MNE/DATA",update_path=false)
limo_epochs

Now we can fit a simple Unfold model to it.

First extract the data & convert it to Julia/Unfold requirements

data = limo_epochs.get_data(picks="B11")
data  = permutedims(data,[2,3,1]) # get into ch x times x epochs

function convert_pandas(df_pd)
      df= DataFrame()
    for col in df_pd.columns
        df[!, col] = getproperty(df_pd, col).values
    end
    return df
end
events = convert_pandas(limo_epochs.metadata)
rename!(events,2=>:coherence) # negative signs in formulas are not good ;)
events.face = string.(events.face) # ugly names, but fast

Next fit an Unfold Model

uf = fit(UnfoldModel,[Any=>(@formula(0~face+coherence),Float64.(limo_epochs.times))],events,data)
results = coeftable(uf)

plot_results(results)

Read some of your own data

We can make use of all PyMNE importer functions to load the data. Try it for your own data! Get starting with Unfold in no-time!

#eeglabdata = PyMNE.io.read_raw_eeglab("pathToEEGLabSet.set")

Contribute?

Some extra conversions are needed to import the data from PyMNE to Unfold (as shown above). We could try putting these in a wrapper function - do you want to tackle this challenge? Would be a great first contribution to the toolbox :-)