import numpy as np
import glob
import sys
from docopt import docopt
import os
import PP
import normalizations as NORM
docstr = """Write something here
Usage:
train.py [options]
Options:
-h, --help Print this message
--mainFolderPath=<str> Main folder path [default: ../../Data/MS2017b/]
--postfix=<str> Postfix of flair images. i.e. to use FLAIR_s postfix is _s. This also determines the train file [default: ]
--pcRange=<str> PC range [default: 1-99]
"""
args = docopt(docstr, version='v0.1')
print(args)
main_folder_path=args['--mainFolderPath']
postfix = args['--postfix']
pc_range = args['--pcRange'].split('-')
save_folder = os.path.join(main_folder_path, 'extra_data/')
save_path = os.path.join(save_folder, 'hist' + postfix + '.txt')
if not os.path.exists(save_folder):
os.makedirs(save_folder)
pc = (int(pc_range[0]), int(pc_range[1]))
m_p = tuple(range(10, 100, 10))
s = (34.681492767333985, 1638.193154296875)
#s = (34, 1638)
def trainLandmarks(main_folder_path = main_folder_path, postfix = postfix):
scan_folders = glob.glob(main_folder_path + 'scans/*')
FLAIR_path = '/pre/FLAIR' + postfix + '.nii.gz'
m_arr = np.zeros([len(scan_folders), len(m_p)])
for i, sf in enumerate(scan_folders):
print "Landmark training: {:4d}/{:4d}\r".format(i, len(scan_folders)),
sys.stdout.flush()
img_str = sf + FLAIR_path
img_np = PP.numpyFromScan(img_str)
p, m = NORM.getLandmarks(img_np)
mapped_m = np.array([int(NORM.mapLandmarks(p, s, x)) for x in m], dtype=np.int64)
m_arr[i, :] = mapped_m
mean_m = np.mean(m_arr, axis = 0, dtype=np.int64)
NORM.writeHistInfo(save_path, pc, s, m_p, mean_m)
#dwi.standardize.write_std_cfg(cfgpath, pc, landmarks, scale, mapped_scores,
# thresholding)
def getScale(main_folder_path = main_folder_path, postfix = postfix):
scan_folders = glob.glob(main_folder_path + 'scans/*')
FLAIR_path = '/pre/FLAIR' + postfix + '.nii.gz'
min_p = None
max_p = None
for i, sf in enumerate(scan_folders):
print "Scale obtaining: {:4d}/{:4d} \r".format(i, len(scan_folders)),
sys.stdout.flush()
img_str = sf + FLAIR_path
img_np = PP.numpyFromScan(img_str)
p, m = NORM.getLandmarks(img_np)
if min_p is None:
min_p = p[0]
max_p = p[1]
if min_p > p[0]:
min_p = p[0]
if max_p < p[1]:
max_p = p[1]
return (min_p, max_p)
'''
def get_stats(pc, scale, landmarks, img, thresholding):
"""Gather info from single image."""
p, scores = dwi.standardize.landmark_scores(img, pc, landmarks,
thresholding)
p1, p2 = p
s1, s2 = scale
mapped_scores = [dwi.standardize.map_onto_scale(p1, p2, s1, s2, x) for x in
scores]
mapped_scores = [int(x) for x in mapped_scores]
return dict(p=p, scores=scores, mapped_scores=mapped_scores)
def train(pc, scale, landmarks, inpaths, cfgpath, thresholding, verbose):
"""Training phase."""
data = []
for inpath in inpaths:
img, _ = dwi.files.read_pmap(inpath)
if img.shape[-1] != 1:
raise Exception('Incorrect shape: {}'.format(inpath))
d = get_stats(pc, scale, landmarks, img, thresholding)
if verbose:
# print(img.shape, dwi.util.fivenum(img), inpath)
# print(d['p'], d['scores'], inpath)
print(img.shape, d['mapped_scores'], inpath)
data.append(d)
mapped_scores = np.array([x['mapped_scores'] for x in data], dtype=np.int)
mapped_scores = np.mean(mapped_scores, axis=0, dtype=mapped_scores.dtype)
mapped_scores = list(mapped_scores)
if verbose:
print(mapped_scores)
dwi.standardize.write_std_cfg(cfgpath, pc, landmarks, scale, mapped_scores,
thresholding)
'''
if __name__ == "__main__":
#obtain the maximum and minimum intensities based on IOI (varies with different pc)
print('Calculating scale...')
s = getScale()
print('Training landmarks...')
trainLandmarks()
print('Done!')
