def main():
pred = predict.predictor()
for name in NAMES:
for tp in TYPES:
# shutil.copyfile(name + '_' + tp + '.json', name + '_' + tp + '_BACKUP.json')
with open(BASE_DIR + name + '_' + tp + '.json', 'r',
newline='') as f:
f_data = json.load(f)
# create a list of the sentiments of each tweet (in order)
sentiment_list = pred.infer(
tweets=list(map(lambda x: x['text'], f_data['statuses'])))
# normalizes the sentiment so that -1 is the most negative and 1 is the most positive
sentiment_list = list(
map(lambda x: (2.0 * x[1]) - 1, sentiment_list))
# add a sentiment field to each tweet and set it to be the normalized value of the sentiment
for (status, sentiment) in zip(f_data['statuses'],
sentiment_list):
status['sentiment'] = sentiment
with open(BASE_DIR + name + '_' + tp + '.json', 'w') as f:
json.dump(f_data, f)
print('Dumped %d lines of data to file' %
len(f_data['statuses']))
def handle_message(event):
text = u'นี้คือ {}'.format(event.message.type)
output = None
if event.message.type == "text":
text = u'แมวววว'
line_bot_api.reply_message(event.reply_token,
TextSendMessage(text=text))
elif event.message.type == "image":
message_content = line_bot_api.get_message_content(event.message.id)
file_name = event.message.id + "_image.jpg"
with open(file_name, 'wb') as fd:
for chunk in message_content.iter_content():
fd.write(chunk)
fd.close()
output = predict.predictor(file_name)
#text = u'สไตล์แวนโกะเลย เมี้ยววววว {}'.format(os.path.join(request.url_root, output))
url_img = os.path.join(request.url_root, output)
if url_img[:5] != 'https':
url_img = 'https' + url_img[4:]
line_bot_api.reply_message(
event.reply_token,
ImageSendMessage(type="image",
original_content_url=url_img,
preview_image_url=url_img))
else:
line_bot_api.reply_message(event.reply_token,
TextSendMessage(text=text))
def main():
if not os.path.isdir(flag.output_dir):
os.mkdir(flag.output_dir)
if flag.mode == 'train':
train_op = train.Trainer(flag)
train_op.train()
elif flag.mode == 'predict':
predict_op = predict.predictor(flag)
predict_op.inference()
elif flag.mode == 'eval':
eval_op = predict.predictor(flag)
eval_op.evaluate()
elif flag.mode == 'cam':
cam_op = predict.predictor(flag)
cam_op.cam()
else:
print 'not supported'
def home_repost():
if request.method == 'POST':
final_symptoms = request.form.getlist('disease_checkbox')
diseases = predictor(final_symptoms)
if request.form.get('accept'):
db.child('Accepted Diseases').child(input_text).update(
{"Symptoms": final_symptoms})
db.child('Accepted Diseases').child(input_text).update(
{"Disease": diseases})
return render_template('disease.html',
text=input_text,
final_symptoms=final_symptoms,
diseases=diseases)
def predict(self):
self.ignore_warning = False
try:
self.pred.should_terminate = True
print("Successfully terminated predicting thread !")
except Exception as err:
print(err)
# KILL EXISTING PROCESS
try:
subprocess.Popen.kill(self.graph_process)
except Exception as err:
print(err)
date = (self.calendar.get_date())
date = datetime.strptime(date,"%Y-%m-%d")
import threading as th
import multiprocessing as mp
daydelta = 0
fill_plot=bool(self.options["fill"].get()) # fill plot
scatter_plot = bool(self.options["scatter"].get()) # use scatter plot
print("plot options",fill_plot,scatter_plot)
use_gpu = bool(self.options["use_gpu"].get()) # use gpu for tensorflow
num_biomass = self.options["num_biomass"].get()
num_biogas = self.options["num_biogas"].get()
num_solar = self.options["num_solar"].get()
biomass_pv = self.options["biomass_pv"].get()
biogas_pv = self.options["biogas_pv"].get()
period = 15
self.pred = predictor(dt=date,model_path=self.model,use_gpu=use_gpu,message_callback=self.power_error_callback) # creat new instance of predictor
self.pred.iteration_delay = self.interval_scaler.get()/1000 # self.predictor loop delay
self.pred.BIOMASS_PV = biomass_pv
self.pred.BIOGAS_PV = biogas_pv
self.pred.num_biomass = num_biomass
self.pred.num_biogas = num_biogas
self.pred.num_solar = num_solar
pred_thread = th.Thread(target=self.pred.run)
pred_thread.start()
self.graph_process = subprocess.Popen([
"python",
"plot_load.py",
"--date",str(date.date()),
"--scatter-plot", str(scatter_plot),
"--fill-plot", str(fill_plot)
])
def home_post():
if request.method == 'POST':
text = request.form.get('symptoms_input')
words = word_extractor(text)
final_symptoms = symptoms(words)
diseases = predictor(final_symptoms)
global input_text
input_text = text
if request.form.get('accept'):
db.child('Accepted Diseases').child(input_text).update(
{"Symptoms": final_symptoms})
db.child('Accepted Diseases').child(input_text).update(
{"Disease": diseases})
return render_template('disease.html',
text=text,
final_symptoms=final_symptoms,
diseases=diseases)
def handle_message(event):
if event.message.type == "text":
line_bot_api.reply_message(event.reply_token,
TextSendMessage(text='แมววว'))
elif event.message.type == "image":
message_content = line_bot_api.get_message_content(event.message.id)
filepath = event.message.id + "image.jpg"
with open(filepath, 'wb') as fd:
for chunk in message_content.iter_content():
fd.write(chunk)
fd.close()
line_bot_api.reply_message(
event.reply_token,
TextSendMessage(text=predict.predictor(filepath)))
else:
line_bot_api.reply_message(
event.reply_token,
TextSendMessage(text="นี้คือ " + event.message.type))
import predict as pd
my_data = [['slashdot', 'USA', 'yes', 18, 'None'],
['google', 'France', 'yes', 23, 'Premium'],
['digg', 'USA', 'yes', 24, 'Basic'],
['kiwitobes', 'France', 'yes', 23, 'Basic'],
['google', 'UK', 'no', 21, 'Premium'],
['(direct)', 'New Zealand', 'no', 12, 'None'],
['(direct)', 'UK', 'no', 21, 'Basic'],
['google', 'USA', 'no', 24, 'Premium'],
['slashdot', 'France', 'yes', 19, 'None'],
['digg', 'USA', 'no', 18, 'None'],
['google', 'UK', 'no', 18, 'None'],
['kiwitobes', 'UK', 'no', 19, 'None'],
['digg', 'New Zealand', 'yes', 12, 'Basic'],
['slashdot', 'UK', 'no', 21, 'None'],
['google', 'UK', 'yes', 18, 'Basic'],
['kiwitobes', 'France', 'yes', 19, 'Basic']]
header = ['Company', 'Country', 'FAQ', 'No. of Visits']
name = raw_input("Company Name:")
country = raw_input("Country :")
faq = raw_input("Visited FAQ (yes/no)")
visits = int(raw_input("No. of Visits"))
tple = [name, country, faq, visits]
pd.predictor(my_data, tple, header, "Subscription Prediction")
import predict as pd
data = [['yes', 'single', 125, 'No'], ['no', 'married', 100, 'No'],
['no', 'single', 70, 'No'], ['yes', 'married', 120, 'No'],
['no', 'divorced', 95, 'Yes'], ['no', 'married', 60, 'No'],
['yes', 'divorced', 220, 'No'], ['no', 'single', 85, 'Yes'],
['no', 'married', 75, 'No'], ['no', 'single', 90, 'Yes']]
header = ["Home Owner", "Marital Status", "Annual Income"]
own = raw_input("Home Owner?(yes/no)")
mar = raw_input("Marital status(single/married/divorced)")
inc = int(raw_input("Annual Income"))
pd.predictor(data, [own, mar, inc], header, "Defaulted Borrower")
import predict
import cv2
weights_path = "../../inceptionV3.299x299.h5"
p = predict.predictor(weights_path)
#print('\n Enter the path to the image\n')
#image_path = input().strip()
#print(p.predict_from_path(image_path))
img = cv2.imread("../../pythonFL/data/test/drawings/0A2FD005-76FF-4C5A-8C81-8179010ED1BB.jpg")
img = cv2.resize(img, (299, 299))
print(p.predict_from_array(img))
