def get_count(self, sf, ROI, CPI, parameters):
glint_CPI = copy.deepcopy(CPI)
preprocess_glint = preprocess(None, sf, glint_CPI, parameters['blur'],
parameters['canny'])
self.H_count = preprocess_glint.g_count(ROI, glint_CPI, parameters,
self.Video)
def get_blur(self, sf, CPI, parameters, ROI_pupil, ROI_glint):
pre_pupil_blur = preprocess(None, sf, CPI, parameters['blur'],
parameters['canny'])
self.pupil_blur = pre_pupil_blur.anal_blur(ROI_pupil, ROI_glint,
self.Video)
def glint_threshold(self, center, sf, CPI, parameters):
pre_glint_threshold = preprocess(center, sf, CPI, parameters['blur'],
parameters['canny'])
return pre_glint_threshold.d_glint()
def pupil_threshold(self, center, sf, CPI, parameters):
pre_pupil_threshold = preprocess(center, sf, CPI, parameters['blur'],
parameters['canny'])
return pre_pupil_threshold.start()
import fileUtils as fu
import logger as lg
parser = argparse.ArgumentParser()
parser.add_argument('-d', '--dset')
parser.add_argument('-a', '--algo')
# parser.add_argument('-r', action='store_true')
args = parser.parse_args()
cfg.config(args.dset, args.algo)
# cfg.config('energydata_complete')
# cfg.config('peugeot_207_01')
lg.initLogger(cfg.ds_name, cfg.algo)
X, y = pp.preprocess('../datasets/' + cfg.ds_name + '.csv', cfg.delimiter, cfg.skiprows, cfg.endColumn, startColumn= cfg.startColumn, targetColumn = cfg.targetColumn, pca = cfg.pca, decimal = cfg.decimal)
train_size = None
if cfg.algo.lower() == 'k-nn':
if cfg.training_set_cap != None:
if X.shape[0] * (1 - cfg.test_size) > cfg.training_set_cap:
train_size = cfg.training_set_cap / X.shape[0]
X_train, X_test, y_train, y_test = train_test_split(X, y, train_size=train_size, test_size=cfg.test_size, random_state=1)
fu.cleanSIDirs('./out/')
if cfg.algo.lower() == 'ann':
import ANN as ann
ann.process(X_train, X_test, y_train, y_test)
elif cfg.algo.lower() == 'k-nn':
import KNN as knn
knn.process(X_train, X_test, y_train, y_test)
model = crnn.CRNN(32, 1, 37, 256)
if torch.cuda.is_available():
model = model.cuda()
print('loading pretrained model from %s' % model_path)
model.load_state_dict(torch.load(model_path))
alphabet = '012345678946066796176100049157000462'
converter = utils.strLabelConverter(alphabet)
transformer = dataset.resizeNormalize((100, 32))
path = input("Enter the path:")
for i in os.listdir(path):
img_path = path + '/' + i
image = preprocess(img_path)
image = Image.open(img_path).convert('L')
image = transformer(image)
if torch.cuda.is_available():
image = image.cuda()
image = image.view(1, *image.size())
image = Variable(image)
model.eval()
preds = model(image)
_, preds = preds.max(2)
preds = preds.transpose(1, 0).contiguous().view(-1)
preds_size = Variable(torch.IntTensor([preds.size(0)]))
sim_pred = converter.decode(preds.data, preds_size.data, raw=False)