def save_to_db():
threading.Timer(60.0, save_to_db).start() # save one data point per minute
if time_to_save_data():
image = util_guru.download(session)
input = util_image.get_single_input_data(image)
output = network.predict(input)[0][0]
d = Data(output)
print "Adding ", time.strftime("%m/%d %H:%M:%S", time.localtime()), " => ", output, "..."
db.session.add(d)
db.session.commit()
def current():
for file in glob.glob("web/static/img/*.jpg"):
os.remove(file)
image = util_guru.download(session)
timestamp = time.time()
file = "web/static/img/img-{}.jpg".format(timestamp)
util_image.save(image, file)
imageFile = "static/img/img-{}.jpg".format(timestamp)
input = util_image.get_single_input_data(image)
output = network.predict(input)[0][0]
return jsonify(currentImg=imageFile, currentNum=output)
mode = raw_input("1: Train\n2: Load\n3: Live\n")
exp = theanets.Experiment(
theanets.feedforward.Regressor,
layers=(config.IMG_W * config.IMG_H, 500, 1)
)
if mode == "1":
trainer.train(exp)
if mode == "2":
exp = exp.load(path="net.data")
print "Manual validation:"
for file in glob.glob("data/manual/*.jpg"):
image = util_image.load(file)
input = util_image.data(image)
input = input.reshape(1, len(input))
output = exp.predict(input)
print "Prediction for ", file, " = ", output
if mode == "3":
exp = exp.load(path="net.data")
session = util_guru.start()
while True:
image = util_guru.download(session)
print "Opening image..."
image.show()
input = util_image.get_single_input_data(image)
output = exp.predict(input)[0][0]
print "Crowded Level: ", output
raw_input("Any key to download another image...")