frites.conn.conn_fcd_corr#
- frites.conn.conn_fcd_corr(conn, roi='roi', times='times', tskip=1, estimator=None, fill_diagonal=nan, dropna=False, verbose=None)[source]#
Compute the correlation on dynamic network.
This function can be used to compute the correlation between time points of a dynamic functional connectivity array, namely :
\[corr(conn_{t_{i}}, conn_{t_{j}})\]- Parameters
- conn
xr.DataArray 3D array of dynamic functional connectivity of shape (n_samples, n_pairs, n_times)
- roi
python:str| ‘roi’ Name of the spatial dimension describing the names of the pairs of brain regions
- times
python:str| ‘times’ Name of the temporal dimension
- tskip
python:int| 1 Number of time point to skip (equivalent to conn[…, ::tskip])
- estimator
frites.estimator|python:None Estimator in order to measure the amount of information shared between two time-series coming from two distinct brain regions. Note that if you want to privide an estimator, be sure that it is made for continuous variables (mi_type=’cc’). By default the correlation is used
- fill_diagonal
python:float|python:None Value to use in order to fill the diagonal. By default, the diagonal is filled with nans
- conn
- Returns
- corr
xr.DataArray The correlation array of shape (n_samples, n_times, n_times)
- corr