P = in_oversized[n, :, :]
P_ = convolve2d(P, dog, 'same')
P_ = P_[left:right, left:right]
# Normalize and mean-free
if options.mf:
P_ -= P_.mean()
if options.norm:
P_max = max(P_.max(), -P_.min())
P_ /= (P_max + 1e-5)
if options.varnorm:
P_var = np.var(P_)
P_ /= (np.sqrt(P_var) + 1e-5)
tbl_out.append("patches", P_)
in_h5.close()
tbl_out.close()
exit(0)
#============================================================
# Safe debug-output
zoom = 6
grid = U.transpose().reshape((D, size, size))
img = tiled_gfs(grid, sym_cm=False, global_cm=True)
img = img.resize((zoom * img.size[0], zoom * img.size[1]))
img.save(out_fname + "-components.png")
grid = P[:100, :, :]
P_max = np.maximum(P_.max(axis=1), -P_.min(axis=1))
print P_max.shape
P_ /= P_max[:, None] + 1e-5
if options.varnorm:
dlog.progress("Normalizing...")
P_var = np.var(P_, axis=1)
P_ /= np.sqrt(P_var[:, None]) + 1e-5
# Make patches to 2d images again...
P_ = P_.reshape((N_patches, size, size))
# And safe resulting patches
for n in xrange(0, N_patches):
if n % batch_size == 0:
dlog.progress("Saving whitened patches (%d)" % n, n / N_patches)
tbl_out.append("patches", P_[n])
# Safe principal components & eigenvalues
for d in xrange(D):
tbl_out.append("eig_vec", U[:, d])
tbl_out.append("eig_val", S[d])
in_h5.close()
tbl_out.close()
exit(0)
#============================================================
# Safe debug-output
zoom = 6
grid = U.transpose().reshape((D, size, size))
P = in_oversized[n,:,:]
P_ = convolve2d(P, dog, 'same')
P_ = P_[left:right, left:right]
# Normalize and mean-free
if options.mf:
P_ -= P_.mean()
if options.norm:
P_max = max(P_.max(), -P_.min())
P_ /= (P_max+1e-5)
if options.varnorm:
P_var = np.var(P_)
P_ /= (np.sqrt(P_var)+1e-5)
tbl_out.append("patches", P_)
in_h5.close()
tbl_out.close()
exit(0)
#============================================================
# Safe debug-output
zoom = 6
grid = U.transpose().reshape( (D, size, size) )
img = tiled_gfs(grid, sym_cm=False, global_cm=True)
img = img.resize( (zoom*img.size[0], zoom*img.size[1]) )
img.save(out_fname+"-components.png")
grid = P[:100,:,:]
P_max = np.maximum(P_.max(axis=1), -P_.min(axis=1))
print P_max.shape
P_ /= P_max[:, None]+1e-5
if options.varnorm:
dlog.progress("Normalizing...")
P_var = np.var(P_, axis=1)
P_ /= np.sqrt(P_var[:, None])+1e-5
# Make patches to 2d images again...
P_ = P_.reshape( (N_patches, size, size) )
# And safe resulting patches
for n in xrange(0, N_patches):
if n % batch_size == 0:
dlog.progress("Saving whitened patches (%d)" % n, n/N_patches)
tbl_out.append("patches", P_[n])
# Safe principal components & eigenvalues
for d in xrange(D):
tbl_out.append("eig_vec", U[:,d])
tbl_out.append("eig_val", S[d])
in_h5.close()
tbl_out.close()
exit(0)
#============================================================
# Safe debug-output
zoom = 6
Q = Q[left:right, left:right]
# Normalize and mean-free
if options.mf:
Q -= Q.mean()
if options.norm:
Q_max = max(Q.max(), -Q.min())
Q /= (Q_max + 1e-5)
if options.varnorm:
Q_var = np.var(Q_)
Q_ /= (np.sqrt(Q_var) + 1e-5)
# Ensure sanity
assert np.isfinite(Q).all()
tbl_out.append("patches", Q)
in_h5.close()
tbl_out.close()
exit(0)
#============================================================
# Safe debug-output
zoom = 6
grid = U.transpose().reshape((D, size, size))
img = tiled_gfs(grid, sym_cm=False, global_cm=True)
img = img.resize((zoom * img.size[0], zoom * img.size[1]))
img.save(out_fname + "-components.png")
grid = P[:100, :, :]
# Normalize and mean-free
if options.mf:
Q -= Q.mean()
if options.norm:
Q_max = max(Q.max(), -Q.min())
Q /= (Q_max+1e-5)
if options.varnorm:
Q_var = np.var(Q_)
Q_ /= (np.sqrt(Q_var)+1e-5)
# Ensure sanity
assert np.isfinite(Q).all()
tbl_out.append("patches", Q)
in_h5.close()
tbl_out.close()
exit(0)
#============================================================
# Safe debug-output
zoom = 6
grid = U.transpose().reshape( (D, size, size) )
img = tiled_gfs(grid, sym_cm=False, global_cm=True)
img = img.resize( (zoom*img.size[0], zoom*img.size[1]) )
img.save(out_fname+"-components.png")
grid = P[:100,:,:]
out_fname = "patches-%d" % size
out_tbl = AutoTable(out_fname+".h5")
images_h5 = tables.openFile(images_fname, "r")
images = images_h5.root.images
N_images = images.shape[0]
#ppi = (N_patches // N_images // 10) + 1
ppi = 4
for n in xrange(N_patches):
if n % 1000 == 0:
dlog.progress("Extracting patch %d" % n, n/N_patches)
if n % ppi == 0:
while True:
img = images[np.random.randint(N_images)]
img = img / img.max()
oversized_batch = pri.extract_patches_from_single_image(img, (oversize, oversize), ppi)
patches_batch = oversized_batch[:, (size//2):(size//2+size), (size//2):(size//2+size)]
variance = np.var( patches_batch.reshape([ppi, -1] ), axis=1)
if np.alltrue(variance > min_var):
break
out_tbl.append('oversized', oversized_batch[n%ppi])
out_tbl.append('patches', patches_batch[n%ppi])
out_tbl.close()
out_tbl = AutoTable(out_fname + ".h5")
images_h5 = tables.openFile(images_fname, "r")
images = images_h5.root.images
N_images = images.shape[0]
#ppi = (N_patches // N_images // 10) + 1
ppi = 4
for n in xrange(N_patches):
if n % 1000 == 0:
dlog.progress("Extracting patch %d" % n, n / N_patches)
if n % ppi == 0:
while True:
img = images[np.random.randint(N_images)]
img = img / img.max()
oversized_batch = pri.extract_patches_from_single_image(
img, (oversize, oversize), ppi)
patches_batch = oversized_batch[:,
(size // 2):(size // 2 + size),
(size // 2):(size // 2 + size)]
variance = np.var(patches_batch.reshape([ppi, -1]), axis=1)
if np.alltrue(variance > min_var):
break
out_tbl.append('oversized', oversized_batch[n % ppi])
out_tbl.append('patches', patches_batch[n % ppi])
out_tbl.close()
