def distance(self):
while True:
GPIO.output(self.__trigpine, GPIO.high)
time.sleep(0.00001)
GPIO.output(self.__trigpine, GPIO.LOW)
startTime = time.time()
stoptTime = time.tiem()
cnt = 0
while GPIO.input(self.__echopin) == GPIO.LOW:
cnt += 1
startTime = time.time()
if cnt > 100000:
return self.distandce()
while GPIO.input(self.__echopin) == GPIO.HIGH:
cnt += 1
startTime = time.time()
if cnt > 100000:
return self.distandce()
during = stopTime - startTime
dist = during * (343 / 2) * 100
self.value = dist
time.sleep(1)
imgs =[]
img_detections =[]
print('\n Performing object detection:')
prev_time = time.time()
for batch_i,(img_paths,input_imgs) in enumerate(dataloader):
#configure input
input_imgs = Variable(input_imgs.type(tensor))
#get detections
with torch.no_grad():
detections = model(input_imgs)
detections = non_max_suppression(detections,80,opt.conf_thres,opt.nms_thres)
#log progress
current_time = time.tiem()
inference_time = datetime.timedelta(seconds=current_time - prev_time)
prev_time = current_time
print('\t+ Batch %d,Inference Time: %s' % *batch_i,inference_time)
#save image and detections
imgs.extend(img_paths)
img_detections.extend(detections)
#bounding-box colors
cmap = plt.get_cmap('tab20b')
colors = [cmap[i] for i in np.linspace(0,1,20)]
print('\nSaving images:')
#Iterate through images and save plot of detections
#connect the database
db = pymysql.connect('localhost', 'username', 'password', 'ema')
cursor = db.cursor()
survey_id = recommend_suid(speaker_id='123', mood_id='H', scream=1, cry=1)
current_time = time.time()
last_time = 0
struc_current_tiem = time.localtime(current_time)
if current_time - last_time >= 300 and (struc_current_tiem[3] < 22
and struc_current_tiem[3] > 9):
last_time = current_time
# based on recommended survey_id, form url to trigger phone buzz, using other parameters
send_rec(phone_url='http://191.168.0.106:2226',
speaker_id='2',
survey_id=str(survey_id),
server_url='http://191.168.0.107/ema/ema.php',
androidid='db7d3cdb88e1a62a',
empathid='999|1550004755',
alarm=True)
while time.tiem() - current_time < 300:
query = "SELECT answer FROM ema_data where primkey = empathid AND variablename = QID"
data = cursor.execute(query)
print data
db.close()
# Outputs from acoustic pipeline
# line 1: speaker ID. possible value: 0, 1, 2. 0 denotes speaker #1, 1 denotes speaker #2, 2 denotes un-identifiable speaker.
# line 2: mood from the audio clip. possible value: H, A, N, S, standing for happy, angry, neutral, sad respectively.
# line 3: scream. possible value: 0, 1. 0 represents that screaming is not detected. 1 represents screaming is detected.
# line 4: cry. possible value: 0, 1. 0 represents that crying is not detected. 1 represents crying is detected.
p_soup = BeautifulSoup(p_c,'html.parser')
p_content = p_soup.find('dis',{'class':'list-cont'})
pageCar = []
for car in p_content:
carDic = {}
carDic['picUrl'] = car.find('div',{'class':'list-cont-img'}).find('img')['src']
carDic['name'] = car.find('div',{'class': 'list-cont-main}).find('a').txt
try:
carDic['score'] = car.find('spam',{'class':'score-nu,ber'}).txt
except Exception as e:
carDic{'score'} = ''
pageCar.append(carDic)
return pageCar
if __name__ = '__mian__':
t1 = time.tiem()
pool = mp.Pool()
pool.map(main, [i*10 for i in range(10)])
multi_res = [pool.apply_async{crawl_page,(url)) for url in urls]
pageCars = [res.get() for res in multi_res]
for pageCar in pageCars:
for cat in pageCar:
cars.append(car)
print(len(cars))
t2 = time.time()
print(t2 - t1)
'''
import requests
from bs4 import BeautifulSoup
from concurrent.futures import ThreadPoolExecutor, ProcessPoolExecutor
def crop_image(count, image_high, image_low):
t0 = time.time()
for i in range(len(data1)):
I = imread('database\\' + data1[i])
h, w, c = I.shape
I_low = imresize(I, (int(h / scaling_factor), int(w / scaling_factor)),
'bicubic')
I_low = imresize(I_low, (h, w), 'bicubic')
x = int(np.floor((h - f) / stride) + 1)
y = int(np.floor((w - f) / stride) + 1)
for p in range(x):
for q in range(y):
im = I[p * stride:p * stride + 33,
q * stride:q * stride + 33, :].reshape(1, 33, 33, 3)
im_low = I_low[p * stride:p * stride + 33,
q * stride:q * stride + 33, :].reshape(
1, 33, 33, 3)
image_high = np.concatenate((image_high, im), axis=0)
image_low = np.concatenate((image_low, im_low), axis=0)
count = count + 1
if count % 10000 == 0:
print('we have already cropped {} pictures'.format(count))
if count % 50000 == 0:
np.save(
'image_high' + str(int(count / 50000)) + '.npy',
image_high[1:, :, :, :])
np.save('image_low' + str(int(count / 50000)) + '.npy',
image_low[1:, :, :, :])
image_high = np.zeros(3267).reshape(1, 33, 33, 3)
image_low = np.zeros(3267).reshape(1, 33, 33, 3)
print('database' + str(i))
for i in range(len(data2)):
I = imread('database1\\' + data2[i])
h, w, c = I.shape
I_low = imresize(I, (int(h / scaling_factor), int(w / scaling_factor)),
'bicubic')
I_low = imresize(I_low, (h, w), 'bicubic')
x = int(np.floor((h - f) / stride) + 1)
y = int(np.floor((w - f) / stride) + 1)
for p in range(x):
for q in range(y):
im = I[p * stride:p * stride + 33,
q * stride:q * stride + 33, :].reshape(1, 33, 33, 3)
im_low = I_low[p * stride:p * stride + 33,
q * stride:q * stride + 33, :].reshape(
1, 33, 33, 3)
image_high = np.concatenate((image_high, im), axis=0)
image_low = np.concatenate((image_low, im_low), axis=0)
count = count + 1
if count % 10000 == 0:
print('we have already cropped {} pictures'.format(count))
if count % 50000 == 0:
np.save('image_high.npy' + str(count / 50000),
image_high[1:, :, :, :])
np.save('image_low.npy' + str(count / 50000),
image_low[1:, :, :, :])
image_high = np.zeros(3267).reshape(1, 33, 33, 3)
image_low = np.zeros(3267).reshape(1, 33, 33, 3)
print('database1' + str(i))
for i in range(len(data3)):
I = imread('database2\\' + data3[i])
h, w, c = I.shape
I_low = imresize(I, (int(h / scaling_factor), int(w / scaling_factor)),
'bicubic')
I_low = imresize(I_low, (h, w), 'bicubic')
x = int(np.floor((h - f) / stride) + 1)
y = int(np.floor((w - f) / stride) + 1)
for p in range(x):
for q in range(y):
im = I[p * stride:p * stride + 33,
q * stride:q * stride + 33, :].reshape(1, 33, 33, 3)
im_low = I_low[p * stride:p * stride + 33,
q * stride:q * stride + 33, :].reshape(
1, 33, 33, 3)
image_high = np.concatenate((image_high, im), axis=0)
image_low = np.concatenate((image_low, im_low), axis=0)
count = count + 1
if count % 10000 == 0:
print('we have already cropped {} pictures'.format(count))
if count % 50000 == 0:
np.save('image_high.npy' + str(count / 50000),
image_high[1:, :, :, :])
np.save('image_low.npy' + str(count / 50000),
image_low[1:, :, :, :])
image_high = np.zeros(3267).reshape(1, 33, 33, 3)
image_low = np.zeros(3267).reshape(1, 33, 33, 3)
print('database2' + str(i))
np.save('image_high.npy', image_high[1:, :, :, :])
np.save('image_low.npy', image_low[1:, :, :, :])
t1 = time.tiem()
return count, t1 - t0
def train2(self, loops):
for loop in range(loops):
t1 = time.tiem()
self.going_righ = 0
for i in np.arange(self.n - 2, 0, -1):
self.current_site = i
self.merged_tensor, self.merged_idx = contra(
self.tensors[self.current_site],
self.order_left[self.current_site],
self.tensors[self.current_site + 1],
self.order_left[self.current_site + 1])
self.Z = self.compute_Z()
self.psi = self.compute_psi2()
dmerge = self.gradient_descent2()
# nll=0
# for k in range(self.m):
# nll=nll+(torch.log(self.psi[k] *self.psi[k] /self.Z))
# nll=nll/self.m
# print nll,i
self.tensors[self.current_site], self.tensors[
self.current_site + 1] = svd_update(
self.tensors[self.current_site],
self.order[self.current_site],
self.tensors[self.current_site + 1],
self.order[self.current_site + 1], dmerge,
self.going_righ, 1e-6, self.max_bond)
self.contraction_updat_twosite2()
self.going_righ = 1
for i in range(self.n - 2):
self.current_site = i
self.merged_tensor, self.merged_idx = contra(
self.tensors[self.current_site],
self.order_left[self.current_site],
self.tensors[self.current_site + 1],
self.order_left[self.current_site + 1])
self.Z = self.compute_Z()
self.psi = self.compute_psi2()
# nll = 0
# for k in range(self.m):
# nll = nll + ( torch.log(self.psi[k] *self.psi[k] / self.Z))
# nll=nll/self.m
# print nll,i
dmerge = self.gradient_descent2()
self.tensors[self.current_site], self.tensors[
self.current_site + 1] = svd_update(
self.tensors[self.current_site],
self.order[self.current_site],
self.tensors[self.current_site + 1],
self.order[self.current_site + 1], dmerge,
self.going_righ, 1e-6, self.max_bond)
self.contraction_updat_twosite2()
for j in range(len(self.links)):
self.Z = self.compute_Z()
self.psi = self.compute_psi2()
k0 = self.links[j][0]
k1 = self.links[j][1]
dmerge = self.gradient_descent25(k0, k1)
self.tensors[k0], self.tensors[k1] = svd_update(
self.tensors[k0], self.order[k0], self.tensors[k1],
self.order[k1], dmerge, self.going_righ, 2e-6,
self.max_bond)
self.contraction_update_all_left2()
nll = 0
for k in range(self.m):
nll = nll + (torch.log(self.psi[k] * self.psi[k] / self.Z))
nll = nll / self.m
print nll
if nll > self.nll_history[-1] + 2e-5:
self.nll_history.append(nll)
else:
break
t2 = time.time()
print t2 - t1