def Loggedin():
if request.method == 'POST':
#patient_id = request.form['patient_id']
age = int(request.form['age'])
sex = int(request.form['sex'])
cp = int(request.form['cp'])
bp = int(request.form['resting_bp'])
cholesterol = int(request.form['serum_cholestrol'])
blood_sugar = int(request.form['fasting_sugar'])
heart_rate = int(request.form['max_heart_rate'])
exercise = int(request.form['exercise'])
global g_bp, g_cholesterol, g_bloodsugar, g_heartrate, g_exercise, g_out
g_bp = bp
g_cholesterol = cholesterol
g_bloodsugar = blood_sugar
g_heartrate = heart_rate
g_exercise = exercise
#blood_sugar=int(blood_sugar)
#print "down "
#dbHandler.insertPatientDetails(patient_id,age,sex,bp,cholesterol,blood_sugar,
# heart_rate,exercise)
#print "down 1"
reload(test)
output = test.prediction(age, sex, cp, bp, cholesterol, blood_sugar,
heart_rate, exercise)
g_out = output
return render_template('products.html', **globals())
else:
return render_template('index.html')
def test():
data = request.get_json()
pred = prediction(data)
response_pickled = jsonpickle.encode(pred)
return Response(response=response_pickled,
status=200,
mimetype="application/json")
def home():
if request.method == 'GET':
return render_template("index.html")
else:
# var_list = ['Pclass','Sex','Age_band','Title','Embarked','Fare_cat','Fare_cat','Alone','Family_Size','SibSp','Parch']
var_list = ['Pclass','Sex','Age_band','Title','Embarked','Fare_cat','SibSp','Parch']
data = {}
for var in var_list:
data[var] = int(request.form[var])
result = t.prediction(data)
return render_template("index.html", result=result)
# Prepare TCNN extractor
print("Prepare TCNN extractor")
conv1_1_weigths = get_conv_1_1_weights(vggCustom_weights_path)
tcnn_bottom = create_tcnn_bottom(vggCustom_weights_path,
conv1_1_weigths)
tcnn_extractor = tcnn_bottom.predict
img_size = 224
elif tcnn_type == 'squeezenet':
tcnn_model = load_model(squeezenet_tcnn_model_path)
# Prepare TCNN extractor
print("Prepare TCNN extractor")
conv1_weights = get_conv1_weights(squeezeNetCustom_weights_path)
tcnn_bottom = create_squeezenet_tcnn_bottom(
squeezeNetCustom_weights_path, conv1_weights)
tcnn_extractor = tcnn_bottom.predict
img_size = 227
client = init(mqttHost="test.mosquitto.org", client_name="JSON", port=1883)
K.get_session().run(tf.global_variables_initializer())
prediction(tcnn_model,
phrnn_model,
tcnn_extractor,
test_file,
client,
img_size=img_size,
tcnn_type=tcnn_type,
phrnn_type="landmarks")
def evaluate_image(image):
return prediction(image)
tcnn_model = load_model(squeezenet_tcnn_model_path)
# Prepare TCNN extractor
print("Prepare TCNN extractor")
conv1_weights = get_conv1_weights(squeezeNetCustom_weights_path)
tcnn_bottom = create_squeezenet_tcnn_bottom(
squeezeNetCustom_weights_path, conv1_weights)
tcnn_extractor = tcnn_bottom.predict
img_size = 227
input_host = "169.254.163.109"
input_port = 5005
width = 640
height = 480
output_client = init(mqttHost="test.mosquitto.org",
client_name="JSON",
port=1883)
K.get_session().run(tf.global_variables_initializer())
prediction(tcnn_model,
phrnn_model,
tcnn_extractor,
output_client,
input_host,
input_port,
width,
height,
img_size=img_size,
tcnn_type=tcnn_type,
phrnn_type="landmarks")