def __init__(self, ticker):
self.ticker = ticker
params = getParams(ticker)
self.mean = params[0][0]
self.var = params[0][1]
self.vol = params[1]
self.price = get_price(self.ticker)
async def consumer(message, socket, socketID):
global cameraParameters
global isSendingVideo
data = json.loads(message)
if data['type'] == 'get camera params':
await socket.send(json.dumps({'type': 'camera params', 'params': params.getParams()}))
elif data['type'] == 'set camera param':
params.setParam(data['name'], data['value'], cameraParameters)
if (data['name'] == 'camera id'):
init_camera()
elif data['type'] == 'send video':
videoClients.add(socketID)
elif data['type'] == 'stop video':
videoClients.remove(socketID)
elif data['type'] == 'get input config':
config = 'ERROR NO CONFIG'
if (os.path.isfile('configs/inputs.txt')):
file = open('configs/inputs.txt', 'r')
config = file.read()
message = json.dumps({'type': 'input config', 'config': config })
await socket.send(message)
elif data['type'] == 'set input config':
# write data['config'] to file
file = open('configs/inputs.txt', 'w')
file.write(data['config'])
file.close()
def train(dataset_dir, augment=False, pretrained_coco=False):
"""Train the model"""
config = FBSConfig()
dataset_train = FBSDataset()
dataset_train.load_data(dataset_dir, subset='train')
dataset_train.prepare()
dataset_val = FBSDataset()
dataset_val.load_data(dataset_dir, subset='validate')
dataset_val.prepare()
augmentation = None
if augment:
augmentation = iaa.SomeOf((0, 6), [
iaa.Fliplr(0.5),
iaa.Flipud(0.5),
iaa.OneOf([
iaa.Affine(rotate=90),
iaa.Affine(rotate=180),
iaa.Affine(rotate=270)
]),
iaa.Multiply((0.8, 1.5)),
iaa.GaussianBlur(sigma=(0.0, 2.0))
])
model = modellib.MaskRCNN(mode='training',
config=config,
model_dir=DEFAULT_MODEL_DIR)
params = getParams('Mask_RCNN')
epochs = params['epochs']
#TODO remove useless callbacks
if pretrained_coco:
COCO_MODEL_PATH = 'mask_rcnn_coco.h5'
model.load_weights(COCO_MODEL_PATH,
by_name=True,
exclude=[
"mrcnn_class_logits", "mrcnn_bbox_fc",
"mrcnn_bbox", "mrcnn_mask"
])
model.train(dataset_train,
dataset_val,
learning_rate=config.LEARNING_RATE,
epochs=10,
augmentation=augmentation,
layers='heads')
model.train(dataset_train,
dataset_val,
learning_rate=config.LEARNING_RATE,
epochs=25,
augmentation=augmentation,
layers='all')
def train(dataset_dir, augment=False, pretrained_coco=False):
"""Train the model"""
config = FBSConfig()
config.display()
dataset_train = FBSDataset()
dataset_train.load_data(dataset_dir, subset='train')
dataset_train.prepare()
dataset_val = FBSDataset()
dataset_val.load_data(dataset_dir, subset='validate')
dataset_val.prepare()
augmentation = None
if augment:
augmentation = iaa.SomeOf((0, 6), [
iaa.Fliplr(0.5),
iaa.Flipud(0.5),
iaa.OneOf([
iaa.Affine(rotate=90),
iaa.Affine(rotate=180),
iaa.Affine(rotate=270)
])
])
model = modellib.MaskRCNN(mode='training',
config=config,
model_dir=DEFAULT_MODEL_DIR)
params = getParams('Mask_RCNN')
epochs = params['epochs']
if pretrained_coco:
COCO_MODEL_PATH = 'mask_rcnn_coco.h5'
model.load_weights(COCO_MODEL_PATH,
by_name=True,
exclude=[
"conv1", "mrcnn_class_logits", "mrcnn_bbox_fc",
"mrcnn_bbox", "mrcnn_mask"
])
# model_path = model.find_last()
# # Load trained weights
# print("Loading weights from ", model_path)
# model.load_weights(model_path, by_name=True)
model.train(dataset_train,
dataset_val,
learning_rate=config.LEARNING_RATE,
epochs=70,
augmentation=augmentation,
save_best_only=True,
monitored_quantity='val_mrcnn_mask_loss',
layers='all')
def getNotifications(type):
params = getParams(type)
collectionRef = db.collection('{}-notifications'.format(type))
account = getAccount(type)
browser = login(type, account)
browser.open(params['notifsUrl'])
soup = browser.parsed
ul = soup.find('ul', attrs={'class': 'timeline'})
if not ul:
raise ValueError("Couldn't open Notifications page")
li_time_label = ul.find_all('li', attrs={'class': 'time-label'})
div_timeline_item = ul.find_all('div', attrs={'class': 'timeline-item'})
for i in range(len(li_time_label)):
date = li_time_label[i].text.strip(' \t\n\r')
time = div_timeline_item[i].span.text.strip(' \t\n\r').replace(
' ', '').replace(';', '')
heading = div_timeline_item[i].find('h4',
attrs={
'class': 'timeline-header'
}).text.strip(' \t\n\r')
body = div_timeline_item[i].find('div',
attrs={
'class': 'timeline-body'
}).text
poster = div_timeline_item[i].find('h3',
attrs={
'class': 'timeline-header up'
}).text.strip(' \t\n\r')
poster = poster[len('Posted by : \n \n '):]
key = '{} : {} : {}'.format(getHash(date + time + heading + body[:5]),
date, time)
docRef = collectionRef.document(key)
if not docRef.get().exists:
docRef.set({
'key': key,
'date': date,
'time': time,
'heading': heading,
'body': body,
'poster': poster
})
sendMessage(label=type,
message=notificationTemplate(date, time, heading, body,
poster))
return '{} notifications parsed'.format(type)
def login(type, account):
params = getParams(type)
browser = RoboBrowser(history=True, parser='html.parser')
browser.open(params['loginUrl'])
form = browser.get_form(0)
if not form:
raise ValueError("Couldn't login")
form[params['username_field']].value = account['username']
form[params['password_field']].value = account['password']
browser.submit_form(form)
browser.open(params['notifsUrl'])
soup = browser.parsed
ul = soup.find('ul', attrs={'class': 'timeline'})
return True if ul else False
def findAccount(type):
params = getParams(type)
branches = [
'CO', 'SE', 'IT', 'EC', 'EL', 'EE', 'CE', 'PS', 'BT', 'EP', 'MC', 'ME',
'AM', 'PE', 'EN'
]
for branch in branches:
for i in range(1, 100):
rollno = '00' + str(i) if i < 10 else '0' + str(i)
account = {
'username': params['year'] + '/' + branch + '/' + rollno,
'password': '******'
}
print("Trying with account: {}".format(account['username']))
if login(type, account):
return account
def login(type, account):
params = getParams(type)
browser = RoboBrowser(history=True, parser='html.parser')
browser.open(params['loginUrl'])
form = browser.get_form(0)
if not form:
raise ValueError("Couldn't login")
form[params['username_field']].value = account['username']
form[params['password_field']].value = account['password']
browser.submit_form(form)
browser.open(params['notifsUrl'])
print('Logged in with account: {}'.format(account))
return browser
def getJobs(type):
params = getParams(type)
collectionRef = db.collection('{}-jobs'.format(type))
account = getAccount(type)
browser = login(type, account)
browser.open(params['jobsUrl'])
soup = browser.parsed
table_jobopenings = soup.find('table', attrs={'id': 'jobs_search'})
if not table_jobopenings:
raise ValueError("Couldn't open jobs page")
trs = table_jobopenings.find_all('tr')
for i in range(1, len(trs)):
tds = trs[i].find_all('td')
if tds[3].find('i')['class'][1] == 'fa-check':
name = tds[0].text
appDeadline = tds[2].text
dateOfVisit = tds[6].text
link = trs[i]['onclick'].replace("void window.open('",
'').replace("')", '')
docRef = collectionRef.document(name)
if not docRef.get().exists:
docRef.set({
'name': name,
'appDeadline': appDeadline,
'dateOfVisit': dateOfVisit,
'link': link,
})
sendMessage(label=type,
message=jobTemplate(name, appDeadline, dateOfVisit,
link))
return '{} jobs parsed'.format(type)
dataset_train = BrainDataset()
dataset_train.load_brain_data('../../data/train/images/*',
'../../data/train/masks/*')
dataset_train.prepare()
dataset_val = BrainDataset()
dataset_val.load_brain_data('../../data/test/images/*',
'../../data/test/masks/*')
dataset_val.prepare()
LOG_DIR = os.path.join(ROOT_DIR, 'logs')
MODEL_DIR = os.path.join(LOG_DIR, "mask_rcnn")
model = modellib.MaskRCNN(mode='training', config=config, model_dir=MODEL_DIR)
params = getParams('Mask_RCNN')
epochs = params['epochs']
Checkpoint, EarlyStop, ReduceLR, Logger, TenBoard = getCallbacks(params)
# COCO_MODEL_PATH = 'mask_rcnn_coco.h5'
# model.load_weights(COCO_MODEL_PATH, by_name=True,
# exclude=["mrcnn_class_logits", "mrcnn_bbox_fc",
# "mrcnn_bbox", "mrcnn_mask"])
# augmentation = iaa.SomeOf((0, 3),[
# iaa.Fliplr(0.5),
# iaa.Flipud(0.5),
# iaa.OneOf([
# iaa.Affine(rotate=90),
# iaa.Affine(rotate=180),
def savePreferences(*args):
params.getParams()
settings = utility.JsonUtility.createJsonData()
utility.JsonUtility.write(storage.prefsFile, settings)
utility.setAllMetadata()
"output": False,
"n": False,
"r": False,
"s": False,
"h": False,
"c": False,
"l": False,
"i": False,
"start": False,
"e": False,
"missing": False,
"all": False,
"padding": False
}
params.getParams(parameters)
params.paramsCheck(parameters)
cols = 0
# promenna ve ktere uchovavam cislo radku
spaces = 0
# promenna ve ktere uchovavam pocet tabulatoru
head = [
] # promenna ve ktere je ulozen prvni radek pokud je aktivni prepinac -h
x = 0 # pocitadlo prvku v hlavicce
rowCount = int(parameters["start"])
#funkce nahradi problematicke znaky
def replace(st):
st = st.replace("&", "&").replace("'", "'").replace(