pyhrf.paradigm module

class pyhrf.paradigm.Paradigm(stimOnsets, sessionDurations=None, stimDurations=None)

delete_condition(cond)

classmethod from_csv(csvFile, delim=None)

Create a Paradigm object from a CSV file which columns are: session, task name, stimulation onset, stimulation duration, [amplitude]

classmethod from_session_dict(d, sessionDurations=None)

classmethod from_spm_mat(spm_mat_file)

TODO: handle session durations

get_info(long=True)

get_joined_and_rastered(dt)

get_joined_durations()

For each condition, join stimulus durations of all sessions.

get_joined_durations_dim()

For each condition, join stimulus durations of all sessions.

get_joined_onsets()

For each condition, join onsets of all sessions.

get_joined_onsets_dim()

For each condition, join onsets of all sessions.

get_nb_trials()

get_rastered(dt, tMax=None)

Return binary sequences of stimulus arrivals. Each stimulus event is approximated to the closest time point on the time grid defined by dt. eg return: { ‘cond1’ : [np.array([ 0 0 0 1 0 0 1 1 1 0 1]),

np.array([ 0 1 1 1 0 0 1 0 1 0 0])] },

‘cond2’

: [np.array([ 0 0 0 1 0 0 1 1 1 0 0]),

np.array([ 1 1 0 1 0 1 0 0 0 0 0])] },

Arg:

• dt (float): temporal resolution of the target grid

• tMax (float): total duration of the paradigm

  If None, then use the session lengths

get_stimulus_names()

get_t_max()

join_sessions()

save_csv(csvFile)

save_spm_mat_for_1st_level_glm(mat_file, session=0)

to_nipy_paradigm()

pyhrf.paradigm.check_stim_durations(stim_onsets, stimDurations)

If no durations specified (stimDurations is None or empty np.array) then assume spiked stimuli: return a sequence of zeros with same shape as onsets sequence. Check that durations have same shape as onsets.

pyhrf.paradigm.contrasts_to_spm_vec(condition_list, contrasts)

pyhrf.paradigm.extend_sampled_events(sampled_events, sampled_durations)

Add events to encode stimulus duration

pyhrf.paradigm.merge_onsets(onsets, new_condition, criterion=None, durations=None, discard=None)

Convention for definition of onsets or durations: OrderedDict({

‘condition_name’: [<array of timings for sess1>,

<array of timings for sess2>, ...]

}

pyhrf.paradigm.restarize_events(events, durations, dt, t_max)

build a binary sequence of events. Each event start is approximated to the nearest time point on the time grid defined by dt and t_max.