def on_text(msg):
try:
u = json.loads(msg['text'])
if 'dbm' in u:
Datastore.db_store_uplink(msg['time'], u['dbm'], u['ip'], u['up'], u['rat'], u['sig'], u['net'])
except ValueError:
pass
def Delete(self, datasegment = None, tagslist = None): if datasegment is not None or tagslist is not None: if datasegment == None: Datastore.DeleteTag(self.cur,tagslist) elif tagslist == None: Datastore.DeleteValue(self.cur,datasegment) else: Datastore.DeleteValueByTags(self.cur,datasegment,tagslist) else: return -1
def store_by_tags_list(self, data_string, tag_list):
if len(data_string) == 0:
msg = {"status" : -1, "message" : "Data string is empty"}
return json.dumps(msg)
if tag_list == None:
msg = {"status" : -2, "message" : "No tags in supplied data"}
return json.dumps(msg)
tag_list = [tag.strip() for tag in tag_list if len(tag.strip()) != 0]
Datastore.update_tag_value(self.cur,data_string,tag_list)
def store_by_tags_string(self, data_string, tags_string, delimiter = " "):
if len(data_string) == 0:
msg = {"status" : -1, "message" : "Data string is empty"}
return json.dumps(msg)
data_string, tag_list = Parser.parse_data_tags(data_string, tags_string, delimiter)
if tag_list == None:
msg = {"status" : -2, "message" : "No tags in supplied data"}
return json.dumps(msg)
Datastore.update_tag_value(self.cur,data_string,tag_list)
def get(self):
#self.response.headers['Content-Type'] = 'text/plain'
#self.response.write('starting parse')
day = datetime.datetime.today().weekday()
if (day == 2) or (day == 4):
script.update_current_semester()
script.update_current_year()
if day == 3:
try:
Datastore.delete_aggr()
except:
pass
script.update_aggregates(1)
if day == 5:
script.update_aggregates(2)
def storeByTagsString(self, data_string, tags_string, delimiter = " "):
if len(data_string) == 0:
msg = {"status" : -1, "message" : "Data string is empty"}
return json.dumps(msg)
data_string, tag_list = Parser.parse_data_tags(data_string, tags_string, delimiter)
if tag_list == None:
msg = {"status" : -2, "message" : "No tags in supplied data"}
return json.dumps(msg)
Datastore.UpdateTagValue(self.cur,data_string,tag_list)
def __init__(self, db_file):
"""
Initializer for the class model. This will want to know which file to look at for the db.
Then it will provide a connection to that database and allow for some ORM opterations, such
as insert and select.
db_file :str
Example Usage:
cm = ClassModel.ClassModel('testing.db')
"""
# init db connection using the datastore
self.conn = Datastore.DB(db_file).ret().conn
def storeByTagsList(self, data_string, tag_list):
if len(data_string) == 0:
msg = {"status" : -1, "message" : "Data string is empty"}
return json.dumps(msg)
if tag_list == None:
msg = {"status" : -2, "message" : "No tags in supplied data"}
return json.dumps(msg)
tag_list = [tag.strip() for tag in tag_list if len(tag.strip()) != 0]
Datastore.UpdateTagValue(self.cur,data_string,tag_list)
print
print "*************"
return tag_list
def __init__(self, db_file):
"""
Initializer for the road maps toggles model. This will want to know which file to look at for the db.
Then it will provide a connection to that database and allow for some ORM opterations, such
as insert and select.
:param
db_file :str
Example Usage:
tm = TogglesModel.TogglesModel('testing.db')
"""
# init db connection using the datastore
self.conn = Datastore.DB(db_file).ret().conn
"""INSERT INTO "main"."roadmap_toggles"("id","requirements_id","students_id","highlight","created_at",
def delete(self, data_segment = None, tagslist = None): if data_segment is not None or tagslist is not None: if data_segment == None: #delete tag Datastore.delete_tag(self.cur, tagslist) elif tagslist == None: #delete a particular data element if type(data_segment) is list: for segment in data_segment: Datastore.delete_value(self.cur, segment) else: Datastore.delete_value(self.cur, data_segment) else: Datastore.delete_value_by_tags(self.cur, data_segment, tagslist) else: return -1
def store(self, data_string):
if len(data_string) == 0:
msg = {"status" : -3, "message" : "Data string is empty"}
return json.dumps(msg)
data_string, tag_list = Parser.parse_tags(data_string)
if tag_list == None:
msg = {"status" : -2, "message" : "No tags in supplied data"}
return json.dumps(msg)
result = Datastore.update_tag_value(self.cur, data_string,tag_list)
if result == -1:
msg = {"status" : -1, "message" : "error"}
return json.dumps(msg)
elif result == 1:
msg = {"status" : 1, "message" : "success"}
return json.dumps(msg)
def on_image(msg):
if msg['type'] == Messaging.ImageMessage.TYPE_PERIODICAL:
filename = Datastore.add_image(msg['src'], msg['time'], msg['data'])
Datastore.db_store_image(msg['src'], msg['time'], filename, len(msg['data']))
elif msg['type'] == Messaging.ImageMessage.TYPE_MOVEMENT:
filename = Datastore.add_image_movement(msg['src'], msg['time'], msg['uuid'], msg['data'])
Datastore.db_store_image_movement(msg['src'], msg['time'], filename, msg['uuid'], len(msg['data']))
# send only the first picture belonging to a group of pictures from a source. uuid is the group identifier
if msg['src'] not in email_alert or email_alert[msg['src']] != msg['uuid']:
email_alert[msg['src']] = msg['uuid']
if not (datetime.time(8, 0) < datetime.datetime.now().time() < datetime.time(15, 0)):
if calendar.timegm(time.gmtime()) > email_alert['last'] + 3600:
email_alert['last'] = calendar.timegm(time.gmtime())
gmail.send('Activity from cam %i' % msg['src'], 'See attachment.', filename)
else:
main_logger.info('skip email alert due to grace period, last alert %u s ago' %
(calendar.timegm(time.gmtime()) - email_alert['last']))
else:
main_logger.info('skip email alert during day')
elif msg['type'] == Messaging.ImageMessage.TYPE_TEST:
filename = Datastore.add_test_image(msg['src'], msg['time'], msg['data'])
main_logger.info('wrote {}'.format(filename))
def cam_on_any(msg):
Datastore.set_variable(msg['src'], 'uptime', msg['uptime'])
main_logger.info('cameras -> local %s' % Messaging.Message.msg_info(msg))
local_messaging.send(msg, serialize=False)
def on_light_control(msg):
Datastore.db_store_light_control(msg['src'], msg['time'], msg['state'], msg['uuid'])
def on_movement(msg):
Datastore.db_store_movement(msg['src'], msg['time'], msg['detector'], msg['state'], msg['uuid'])
def on_variable(msg):
if msg['name'] == 'temperature':
Datastore.db_store_temperature(msg['src'], msg['time'], msg['value'])
else:
Datastore.set_variable(msg['src'], msg['name'], msg['value'])
import Datastore import datetime now = datetime.datetime.now() Datastore.store_image_meta(now, 2, '/data/image.jpg', 12312)
import threading
import Datastore as data
obj = data.Datastore()
threads = []
for _ in range(1):
t = threading.Thread(target=obj.create,
args=['tirumala', '{name:tirumala,age:21}', 1000])
t.start()
threads.append(t)
for thread in threads:
thread.join()
threads = []
for _ in range(1):
t = threading.Thread(target=obj.read, args=['tirumala'])
t.start()
threads.append(t)
for thread in threads:
thread.join()
threads = []
for _ in range(3):
t = threading.Thread(target=obj.delete, args=['tirumala'])
t.start()
threads.append(t)
for thread in threads:
thread.join()
Samuel Lundquist
Priority credit to:
Ryan Gurnick - 3/4/20 Creation
"""
import os
import ClassParser
import Datastore
DB = 'seed.db'
if not os.path.exists('seed.db'):
# setup datastore
ds = Datastore.DB(DB)
# generate them default tables
ds.generateTables()
# all of the subject codes
subject_codes = ["AAAP", "AAD", "ACTG", "AEIS", "AFR", "AIM", "ANTH", "ANTM", "ARB", "ARCH", "ARH", "ART", "ARTC",
"ARTD", "ARTF", "ARTM", "ARTO", "ARTP", "ARTR", "ARTS", "ASIA", "ASL", "ASTR", "BA", "BI", "BIKC",
"BIOE", "BLST", "CARC", "CAS", "CDS", "CFT", "CH", "CHKC", "CHN", "CHNF", "CINE", "CIS", "CIT", "CLAS",
"COLT", "CPSY", "CRDG", "CRES", "CRWR", "CSCH", "DAN", "DANC", "DANE", "DIST", "DSGN", "EALL", "EC",
"ECE", "EDLD", "EDST", "EDUC", "ENG", "ENVS", "ERTH", "ES", "ESC", "EURO", "FHS", "FIN", "FINN", "FLR",
"FR", "GEOG", "GEOL", "GER", "GRK", "GRST", "GSAE", "GSCL", "GSGE", "GSST", "HBRW", "HC", "HIST",
"HPHY", "HUM", "IARC", "ICH", "INTL", "IST", "ITAL", "J", "JDST", "JGS", "JPN", "KC", "KRN", "LA",
"LAS", "LAT", "LAW", "LEAD", "LERC", "LIB", "LING", "LT", "MATH", "MDVL", "MENA", "MGMT", "MIL",
"MKTG", "MUE", "MUJ", "MUP", "MUS", "NAS", "NORW", "OBA", "OIMB", "OLIS", "PD", "PDX", "PE", "PEAQ",
"PEAS", "PEC", "PEF", "PEI", "PEIA", "PEL", "PEMA", "PEMB", "PEO", "PERS", "PERU", "PETS", "PEW",
"PHIL", "PHKC", "PHYS", "PORT", "PPPM", "PREV", "PS", "PSY", "QST", "REES", "REL", "RL", "RUSS",
def main(mainpath, load=False, training=True, weights=False, rgb=0):
torch.cuda.device(0)
plt.ion()
# If data is multi or single channel
if rgb:
tforms = transforms.Compose([transforms.ToTensor(),
transforms.Normalize((0.5, 0.5, 0.5), (0.5, 0.5, 0.5))])
net = UNet(n_channels=3, n_classes=1)
else:
tforms = transforms.Compose([transforms.ToTensor(), transforms.Normalize(0.5, 0.5)])
net = UNet(n_channels=1, n_classes=1)
if training:
trainpath = os.path.join(mainpath, 'image')
filelist = os.listdir(trainpath)
trainmasks = os.path.join(mainpath, 'label')
masklist = os.listdir(trainmasks)
if weights:
if os.path.isdir(os.path.join(mainpath, 'weights')) != 1:
os.mkdir(os.path.join(mainpath, 'weights'))
print("generating weights")
for file in masklist:
img = Image.open(os.path.join(mainpath, 'label', file))
weights = Datastore.generateWeights(img)
weights = Image.fromarray(weights)
weights.save(os.path.join(mainpath, 'weights', file[:-4]+'.tif'))
print("generated weights")
weightspath = os.path.join(mainpath, 'weights')
weightslist = os.listdir(weightspath)
else:
weightspath = os.path.join(mainpath,'weights')
weightslist = os.listdir(weightspath)
dataset = Datastore.Datastore(filelist, masklist, weightslist, mainpath, transforms=tforms)
batch_N = 1
trainloader = torch.utils.data.DataLoader(dataset, batch_size=batch_N, shuffle=True, num_workers=0)
N_train = len(dataset)
gpu = 0
startepoch = 0
if gpu == 1:
gpu = torch.device("cuda:0")
print("Connected to device: ", gpu)
net = net.to(gpu)
epochs = 50
lr = 0.001
val_percent = 0.05
optimizer = optim.SGD(net.parameters(),
lr=lr,
momentum=0.9)
criterion = nn.BCEWithLogitsLoss()
fig = plt.figure(figsize=(18, 5), dpi=80, facecolor='w', edgecolor='k')
fig.tight_layout()
# Load in previous model
if load:
try:
checkpoint = torch.load('model2.pt')
net.load_state_dict(checkpoint['model_state_dict'])
optimizer.load_state_dict(checkpoint['optimizer_state_dict'])
startepoch = checkpoint['epoch'] + 1
loss = checkpoint['loss']
except FileNotFoundError:
print(f"No model file found at {mainpath}")
train(net, optimizer, criterion, trainloader, startepoch, epochs, gpu, batch_N, N_train, mainpath)
else:
checkpoint = torch.load('model2.pt')
net.load_state_dict(checkpoint['model_state_dict'])
predict(net, mainpath)
