def generate_training_data_vol(jobs, case):
print("Generating Training Data (Volume)...")
images_fn, labels_fn, regint_fn = get_filenames(path, im_name, lb_name, ro_name)
kernels = generate_kernels()
for i in range(len(images_fn)):
if i == case:
create_volume_info(images_fn, labels_fn, regint_fn, kernels, i, jobs)
break
def generate_training_data_vol(jobs, case):
print("Generating Training Data (Volume)...")
images_fn, labels_fn, regint_fn = get_filenames(path, im_name, lb_name,
ro_name)
kernels = generate_kernels()
for i in range(len(images_fn)):
if i == case:
create_volume_info(images_fn, labels_fn, regint_fn, kernels, i,
jobs)
break
def generate_training_data(jobs):
print("Generating Training Data...")
slice_infos = []
images_fn, labels_fn, regint_fn = get_filenames(path, im_name, lb_name, ro_name)
kernels = generate_kernels()
slice_infos = Parallel(n_jobs=jobs)(delayed(create_sliceinfo_w)(images_fn,
labels_fn, regint_fn, kernels, i) for i in range(len(images_fn)))
with open(pickle, 'wb') as f:
dill.dump(slice_infos, f)
def generate_training_data(jobs):
print("Generating Training Data...")
slice_infos = []
images_fn, labels_fn, regint_fn = get_filenames(path, im_name, lb_name, ro_name)
kernels = generate_kernels()
slice_infos = Parallel(n_jobs=jobs)(delayed(create_sliceinfo_w)(images_fn,
labels_fn, regint_fn, kernels, i) for i in range(len(images_fn)))
with open(pickle, 'wb') as f:
dill.dump(slice_infos, f)
def bigtest():
# really long test to generate SliceInfos at all radiuses
images_fn, labels_fn = get_filenames(path, im_name, lb_name)
# go through all candidate psizes
for p in psizes:
global psize
psize = p
t0 = time()
slice_infos = create_sliceinfos_w(images_fn, labels_fn)
dt = time() - t0
print("Finished in %.2f seconds." % dt)
# save the slice info
with open("slice_infos_" + str(psize) + ".pkl", 'wb') as f:
dill.dump(slice_infos, f)
def run():
if generate:
t0 = time()
# get lists of files to process
images_fn, labels_fn = get_filenames(path, im_name, lb_name)
# generate sliceinfos for all those images
if generate == 1:
slice_infos = create_sliceinfos(images_fn, labels_fn)
elif generate == 2:
slice_infos = create_sliceinfos_w(images_fn, labels_fn)
print(slice_infos[0].vals_m[:30])
dt = time() - t0
print("Finished in %.2f seconds." % dt)
# save the slice info
with open(pickle, 'wb') as f:
dill.dump(slice_infos, f)
else:
with open(pickle, 'rb') as f:
slice_infos = dill.load(f)
total_successes = 0
total_trials = 0
if classify == 1: # Go through each slice and classify PCPs
# PCP = principal component patches
t0 = time()
if len(sys.argv) >= 2:
res = Parallel(n_jobs=int(sys.argv[1]))(
delayed(test_rf_feats)(slice_infos, i)
for i in range(len(slice_infos)))
for successes, trials in res:
total_successes += successes
total_trials += trials
else:
for i in range(len(slice_infos)):
successes, trials = test_rf_feats(slice_infos, i)
total_successes += successes
total_trials += trials
print("Rate: {}/{} = {:.2f}".format(
total_successes, total_trials,
float(total_successes) / float(total_trials) * 100))
dt = time() - t0
print("Took %.2f seconds." % dt)
elif classify == 2: # check classification of all patches for one slice
test_rf_feats_fullspec(slice_infos, fullspec_i)
pass
elif classify == 3: # fully reconstruct for one slice
rf_reconstruct(slice_infos, fullspec_i)
elif classify == 4: # show the reconstruct for one slice
with open(recons, 'rb') as f:
recons_im = dill.load(f)
real_lb = slice_infos[fullspec_i].slice_lb
# for threshval in range(0, 101):
# threshval = threshval / 100.0
# recons_th = threshold(recons_im, threshval, True)
# print("%.2f" % threshval, "%.2f" % compare_im(recons_th, real_lb))
recons_th = threshold(recons_im, 0.45, True)
compare_im(recons_th, real_lb, True)
