lplo = LeavePLabelOut(session_id_fmri, p=2)
for train_index, test_index in lplo:
# Split into train and test sets
fmri_train, fmri_test = fmri[train_index], fmri[test_index]
design_train, design_test = design[train_index], design[test_index]
stimuli_train, stimuli_test = stimuli[train_index], stimuli[test_index]
# Feature selection
fmri_train, fmri_test = de.feature_selection(
fmri_train, fmri_test, np.argmax(stimuli_train, axis=1))
# Fit a ridge regression to predict the design matrix
prediction_test, prediction_train, score = de.fit_ridge(
fmri_train,
fmri_test,
design_train,
design_test,
double_prediction=True,
extra=fmri_train)
# Fit a logistic regression for deconvolution
accuracy = de.logistic_deconvolution(prediction_train,
prediction_test,
stimuli_train,
stimuli_test,
logistic_window,
delay=delay)
scores.append(accuracy)
subjects.append(subject)
models.append('logistic deconvolution')
continue
labels = np.hstack([[session] * 20 for session in range(4)])
lplo2 = LeavePLabelOut(labels, p=1)
for train_id, test_id in lplo2:
fmri_train, fmri_test = fmri_isi[train_id], fmri_isi[test_id]
design_train, design_test = (design_isi[train_id],
design_isi[test_id])
stimuli_train, stimuli_test = (stimuli_isi[train_id],
stimuli_isi[test_id])
# Feature selection
fmri_train, fmri_test = de.feature_selection(
fmri_train, fmri_test, np.argmax(stimuli_train, axis=1))
# Fit a ridge regression to predict the design matrix
prediction_test, prediction_train, score = de.fit_ridge(
fmri_train, fmri_test, design_train, design_test,
double_prediction=True, extra=fmri_train)
# Fit a logistic regression for deconvolution
accuracy = de.logistic_deconvolution(
prediction_train, prediction_test, stimuli_train,
stimuli_test, logistic_window, delay=delay)
scores.append(accuracy)
subjects.append(subject + 1)
models.append('logistic deconvolution')
isis.append(isi)
print('finished subject ' + str(subject))
isi = 4.8
time_window = 3
delay = 3
else:
continue
# Time window + feature selection
fmri_train, fmri_test, stimuli_train, stimuli_test = de.apply_time_window(
fmri_train,
fmri_test,
stimuli_train,
stimuli_test,
delay=delay,
time_window=time_window,
k=k)
# Fit ridge regression
prediction, score = de.fit_ridge(fmri_train, fmri_test, stimuli_train,
stimuli_test)
# Fit a logistic regression for deconvolution
accuracy = de.ridge_scoring(prediction, stimuli_test)
scores.append(accuracy)
subjects.append(subject + 1)
models.append('time-delayed embedding')
isis.append(isi)
print('finished subject ' + str(subject))
for session in range(len(fmri))]
# Stack the BOLD signals and the design matrices
fmri = np.vstack(fmri)
stimuli = np.vstack(stimuli)
session_id_fmri = np.hstack(session_id_fmri)
lplo = LeavePLabelOut(session_id_fmri, p=1)
for train_index, test_index in lplo:
# Split into train and test sets
fmri_train, fmri_test = fmri[train_index], fmri[test_index]
stimuli_train, stimuli_test = stimuli[train_index], stimuli[test_index]
# Time window + feature selection
fmri_train, fmri_test, stimuli_train, stimuli_test = de.apply_time_window(
fmri_train, fmri_test, stimuli_train, stimuli_test, delay=delay,
time_window=time_window, k=k)
# Fit ridge regression
prediction, score = de.fit_ridge(fmri_train, fmri_test, stimuli_train,
stimuli_test)
# Fit a logistic regression for deconvolution
accuracy = de.ridge_scoring(prediction[:, 1:], stimuli_test[:, 1:])
scores.append(accuracy)
subjects.append(subject + 1)
models.append('time-delayed embedding')
print('finished subject ' + str(subject))
stimuli = np.vstack(stimuli)
session_id_fmri = np.hstack(session_id_fmri)
lplo = LeavePLabelOut(session_id_fmri, p=2)
for train_index, test_index in lplo:
# Split into train and test sets
fmri_train, fmri_test = fmri[train_index], fmri[test_index]
design_train, design_test = design[train_index], design[test_index]
stimuli_train, stimuli_test = stimuli[train_index], stimuli[test_index]
# Feature selection
fmri_train, fmri_test = de.feature_selection(
fmri_train, fmri_test, np.argmax(stimuli_train, axis=1))
# Fit a ridge regression to predict the design matrix
prediction_train, prediction_test, score = de.fit_ridge(
fmri_train, fmri_test, design_train, design_test)
# Fit a logistic regression for deconvolution
accuracy = de.logistic_deconvolution(prediction_train, prediction_test,
stimuli_train, stimuli_test,
logistic_window)
scores.append(accuracy)
subjects.append(subject)
print('finished subject ' + str(subject))
dict = {}
dict['subject'] = subjects
dict['accuracy'] = scores