Name

db_hrf - Convolve DNN activation with SPM canonical hemodynamic response function. And match it with the time points of Brain activation.

Synopsis

db_hrf [-h] -act Activation [-layer Layer [Layer ...]]
       [-chn Channel [Channel ...]] [-dmask DnnMask] -stim Stimulus
       -tr TR -n_vol N_volume [-ops Ops] -out Output

Arguments

Required Arguments

Argument	Discription
act	DNN activation file
stim	a .stim.csv file which contains onsets and durations
tr	repetition time of BOLD signal acquisition
n_vol	the number of volumes of BOLD signal
out	output filename with suffix as .act.h5

Optional Arguments

Argument	Discription
layer	layer names of interest
chn	channel numbers of interest Default using all channels of each layer specified by -layer.
dmask	a .dmask.csv file in which layers of interest are listed with their own channels, rows and columns of interest.
ops	choices=(10, 100, 1000) oversampling number per second (default: 100)

Outputs

The output is a .act.h5 file.

Examples

Convolve activation of layer fc3 in test.act.h5 with SPM canonical hemodynamic response function according to onsets and durations in test.stim.csv. And match it with the 152 time points of Brain activation with repetition time as 2 seconds. Save results to out.act.h5.

db_hrf -act test.act.h5 -layer fc3 -stim test.stim.csv -tr 2 -n_vol 152 -out out.act.h5