elephant.spade.concept_output_to_patterns

elephant.spade.concept_output_to_patterns(concepts, winlen, bin_size, pv_spec=None, spectrum='#', t_start=array(0.) * ms)

Construction of dictionaries containing all the information about a pattern starting from a list of concepts and its associated pvalue_spectrum.

Parameters:

conceptstuple

Each element of the tuple corresponds to a pattern which it turn is a tuple of (spikes in the pattern, occurrences of the patterns)

winlenint

Length (in bins) of the sliding window used for the analysis.

bin_sizepq.Quantity

The time precision used to discretize the spiketrains (binning).

pv_specNone or tuple

Contains a tuple of signatures and the corresponding p-value. If equal to None all p-values are set to -1.

spectrum{‘#’, ‘3d#’}

‘#’: pattern spectrum using the as signature the pair:

(number of spikes, number of occurrences)

‘3d#’: pattern spectrum using the as signature the triplets:

(number of spikes, number of occurrence, difference between last and first spike of the pattern)

Default: ‘#’

t_startpq.Quantity

t_start of the analyzed spike trains

Returns:

outputlist

List of dictionaries. Each dictionary corresponds to a pattern and has the following entries:

‘itemset’:

A list of the spikes in the pattern, expressed in theform of itemset, each spike is encoded by spiketrain_id * winlen + bin_id.

‘windows_ids’:

The ids of the windows in which the pattern occurred in discretized time (given byt the binning).

‘neurons’:

An array containing the idx of the neurons of the pattern.

‘lags’:

An array containing the lags (integers corresponding to the number of bins) between the spikes of the patterns. The first lag is always assumed to be 0 and corresponds to the first spike.

‘times’:

An array containing the times (integers corresponding to the bin idx) of the occurrences of the patterns.

‘signature’:

A tuple containing two integers (number of spikes of the patterns, number of occurrences of the pattern).

‘pvalue’:

The p-value corresponding to the pattern. If n_surr==0, all p-values are set to -1.