def layer_ridge_regression(train_feat,
train_fmri,
val_feat,
val_fmri,
out_dir,
prefix,
with_wt=True):
"""Calculate ridge regression between features from one layer and
the fmri responses from all voxels.
"""
train_feat = train_feat.reshape(-1, train_feat.shape[3])
val_feat = val_feat.reshape(-1, val_feat.shape[3])
voxel_size = train_fmri.shape[0]
corr_file = os.path.join(out_dir, prefix + '_corr.npy')
wt, corr, valphas, bscores, valinds = ridge.bootstrap_ridge(
train_feat.T,
train_fmri.T,
val_feat.T,
val_fmri.T,
alphas=np.logspace(-2, 2, 20),
nboots=5,
chunklen=100,
nchunks=10,
single_alpha=True)
np.save(corr_file, corr)
if with_wt:
wt_file = os.path.join(out_dir, prefix + '_weights.npy')
np.save(wt_file, wt)
def cnn_pred_sugar(train_fmri, train_feat, val_fmri, val_feat, corr_mtx, idx):
"""Sugar function for cnn activation prediction."""
row, col = idx[0], idx[1]
print 'row %s - col %s' % (row, col)
tfeat = train_feat[:, row, col, :]
vfeat = val_feat[:, row, col, :]
wt, corr, valphas, bscores, valinds = ridge.bootstrap_ridge(
train_fmri.T,
tfeat.T,
val_fmri.T,
vfeat.T,
alphas=np.logspace(-2, 2, 20),
nboots=5,
chunklen=100,
nchunks=10,
single_alpha=False)
corr_mtx[:, row, col] = corr
def ridge_sugar(train_feat, train_fmri, val_feat, val_fmri, corr_mtx, idx):
"""Sugar function for ridge regression."""
pixel_size = (train_feat.shape[1], train_feat.shape[2])
row, col = idx[0], idx[1]
print 'row %s - col %s' % (row, col)
tfeat = train_feat[:, row, col, :]
vfeat = val_feat[:, row, col, :]
wt, corr, valphas, bscores, valinds = ridge.bootstrap_ridge(
tfeat.T,
train_fmri.T,
vfeat.T,
val_fmri.T,
alphas=np.logspace(-2, 2, 20),
nboots=5,
chunklen=100,
nchunks=10,
single_alpha=True)
corr_mtx[row * pixel_size[0] + col] = corr
def ridge_fitting(feat_dir, prf_dir, db_dir, subj_id, roi):
"""pRF model fitting using ridge regression.
90% trainning data used for model tuning, and another 10% data used for
model seletion.
"""
# load fmri response
vxl_idx, train_fmri_ts, val_fmri_ts = dataio.load_vim1_fmri(db_dir,
subj_id,
roi=roi)
del val_fmri_ts
print 'Voxel number: %s' % (len(vxl_idx))
# load candidate models
train_models = np.load(os.path.join(feat_dir, 'train_candidate_model.npy'),
mmap_mode='r')
# output directory config
roi_dir = os.path.join(prf_dir, roi)
check_path(roi_dir)
# model seletion and tuning
ALPHA_NUM = 20
BOOTS_NUM = 15
paras_file = os.path.join(roi_dir, 'reg_paras.npy')
paras = np.memmap(paras_file,
dtype='float64',
mode='w+',
shape=(42500, len(vxl_idx), 72))
mcorr_file = os.path.join(roi_dir, 'reg_model_corr.npy')
mcorr = np.memmap(mcorr_file,
dtype='float64',
mode='w+',
shape=(42500, len(vxl_idx)))
alphas_file = os.path.join(roi_dir, 'reg_alphas.npy')
alphas = np.memmap(alphas_file,
dtype='float64',
mode='w+',
shape=(42500, len(vxl_idx)))
# fMRI data z-score
print 'fmri data temporal z-score'
m = np.mean(train_fmri_ts, axis=1, keepdims=True)
s = np.std(train_fmri_ts, axis=1, keepdims=True)
train_fmri_ts = (train_fmri_ts - m) / (1e-10 + s)
# split training dataset into model tunning set and model selection set
tune_fmri_ts = train_fmri_ts[:, :int(1750 * 0.9)]
sel_fmri_ts = train_fmri_ts[:, int(1750 * 0.9):]
# model testing
for i in range(42500):
print 'Model %s' % (i)
# remove models which centered outside the 20 degree of visual angle
xi = (i % 2500) / 50
yi = (i % 2500) % 50
x0 = np.arange(5, 500, 10)[xi]
y0 = np.arange(5, 500, 10)[yi]
d = np.sqrt(np.square(x0 - 250) + np.square(y0 - 250))
if d > 249:
print 'Model center outside the visual angle'
paras[i, ...] = np.NaN
mcorr[i] = np.NaN
alphas[i] = np.NaN
continue
train_x = np.array(train_models[i, ...]).astype(np.float64)
train_x = zscore(train_x.T).T
# split training dataset into model tunning and selection sets
tune_x = train_x[:int(1750 * 0.9), :]
sel_x = train_x[int(1750 * 0.9):, :]
wt, r, alpha, bscores, valinds = ridge.bootstrap_ridge(
tune_x,
tune_fmri_ts.T,
sel_x,
sel_fmri_ts.T,
alphas=np.logspace(-2, 3, ALPHA_NUM),
nboots=BOOTS_NUM,
chunklen=175,
nchunks=1,
single_alpha=False,
use_corr=False)
paras[i, ...] = wt.T
mcorr[i] = r
alphas[i] = alpha
# save output
paras = np.array(paras)
np.save(paras_file, paras)
mcorr = np.array(mcorr)
np.save(mcorr_file, mcorr)
alphas = np.array(alphas)
np.save(alphas_file, alphas)
def pls_ridge_fitting(feat_dir, prf_dir, db_dir, subj_id, roi):
"""pRF model fitting using ridge regression.
90% trainning data used for model tuning, and another 10% data used for
model seletion.
"""
# load fmri response
fmri_data = np.load(os.path.join(prf_dir, roi, 'pls_residual_fmri.npz'))
vxl_idx = fmri_data['vxl_idx']
train_fmri_ts = fmri_data['pls_train_residual']
print 'Voxel number: %s' % (len(vxl_idx))
# load candidate models
train_models = np.load(os.path.join(feat_dir, 'train_candidate_model.npy'),
mmap_mode='r')
# output directory config
roi_dir = os.path.join(prf_dir, roi, 'pls_residual')
check_path(roi_dir)
# model seletion and tuning
ALPHA_NUM = 20
BOOTS_NUM = 15
paras_file = os.path.join(roi_dir, 'reg_paras.npy')
paras = np.memmap(paras_file,
dtype='float64',
mode='w+',
shape=(15360, len(vxl_idx), 46))
mcorr_file = os.path.join(roi_dir, 'reg_model_corr.npy')
mcorr = np.memmap(mcorr_file,
dtype='float64',
mode='w+',
shape=(15360, len(vxl_idx)))
alphas_file = os.path.join(roi_dir, 'reg_alphas.npy')
alphas = np.memmap(alphas_file,
dtype='float64',
mode='w+',
shape=(15360, len(vxl_idx)))
# fMRI data z-score
print 'fmri data temporal z-score'
m = np.mean(train_fmri_ts, axis=1, keepdims=True)
s = np.std(train_fmri_ts, axis=1, keepdims=True)
train_fmri_ts = (train_fmri_ts - m) / (1e-10 + s)
# split training dataset into model tunning set and model selection set
tune_fmri_ts = train_fmri_ts[:, :int(7200 * 0.9)]
sel_fmri_ts = train_fmri_ts[:, int(7200 * 0.9):]
# model testing
for i in range(15360):
print 'Model %s' % (i)
train_x = np.array(train_models[i, ...]).astype(np.float64)
train_x = zscore(train_x).T
# split training dataset into model tunning and selection sets
tune_x = train_x[:int(7200 * 0.9), :]
sel_x = train_x[int(7200 * 0.9):, :]
wt, r, alpha, bscores, valinds = ridge.bootstrap_ridge(
tune_x,
tune_fmri_ts.T,
sel_x,
sel_fmri_ts.T,
alphas=np.logspace(-2, 3, ALPHA_NUM),
nboots=BOOTS_NUM,
chunklen=720,
nchunks=1,
single_alpha=False,
use_corr=False)
paras[i, ...] = wt.T
mcorr[i] = r
alphas[i] = alpha
# save output
paras = np.array(paras)
np.save(paras_file, paras)
mcorr = np.array(mcorr)
np.save(mcorr_file, mcorr)
alphas = np.array(alphas)
np.save(alphas_file, alphas)
