def load_images_only(image_paths):
images = []
print("Images:")
for i, image_path in enumerate(image_paths):
print("\t#{}: {}".format(i, image_path))
image, raw_image = preprocess(image_path)
images.append(image)
return images
import gym
from main import stack_frames, preprocess
from model import PolicyGradientAgent
agent = PolicyGradientAgent(learning_rate=0.001,
discount_factor=0.9,
num_actions=6,
chkpt_dir='tmp/checkpoint')
agent.load_checkpoint()
env = gym.make('SpaceInvaders-v0')
observation = env.reset()
observation = preprocess(observation)
stack_size = 4
stacked_frames = None
stacked_frames = stack_frames(stacked_frames, observation, stack_size)
done = False
while not done:
env.render()
action = agent.choose_action(stacked_frames)
obseravtion, reward, done, info = env.step(action)
observation = preprocess(observation)
stacked_frames = stack_frames(stacked_frames, observation, stack_size)
env.close()
def display_graph(self):
self.algorithm = self.combobox_algs.get()
self.mode = self.combobox_types.get()
print(self.mode)
print(self.algorithm)
if self.mode == "Google":
self.file_path = "googl.us.txt"
elif self.mode == "Apple":
self.file_path = "aapl.us.txt"
elif self.mode == "Amazon":
self.file_path = "amzn.us.txt"
elif self.mode == "Coca Cola":
self.file_path = "ko.us.txt"
# df = load_file(self.file_path)
df = pd.read_csv(self.file_path)
# if self.algorithm == "SVM":
# print("Svm izabran")
# # self.y_train, self.y_val, self.y_predict = svm_prediction(df)
#
# new_df = preprocess(load_file(self.file_path))
# x_train, y_train, x_valid, y_valid, self.train, self.valid = train_valid_split(new_df)
# self.y_predict = svm_prediction(df, x_train, y_train, x_valid, y_valid)
# self.y_train = y_train
# self.y_val = y_valid
if self.algorithm == "Moving average":
print("MA izabran")
#NISAM TESTIRAO MA, TREBA MODIFIKOVATI
new_df = preprocess(load_file(self.file_path))
x_train, y_train, x_valid, y_valid, self.train, self.valid = train_valid_split(new_df)
self.y_predict = predict_ma(df)
self.y_train = y_train
self.y_val = y_valid
elif self.algorithm == "KNN":
print("KNN izabran")
df = load_file(self.file_path)
new_df = preprocess(df)
x_train, y_train, x_valid, y_valid, self.train, self.valid = train_valid_split(new_df)
self.y_predict = knn_predict(x_train, y_train, x_valid)
self.y_train = y_train
self.y_val = y_valid
plt.plot(self.y_train)
plt.plot(self.y_val)
plt.plot(self.y_predict)
plt.show()
# plot_graph(self.train, self.valid, self.y_predict)
elif self.algorithm == "Auto Arima":
print("Auto Arima izabran")
# UBACI OVDE POZIV AUTO ARIMA METODE
df = load_file(self.file_path)
new_df = preprocess(df)
x_train, y_train, x_valid, y_valid, self.train, self.valid = train_valid_split(new_df)
self.y_predict = auto_arima_predict(df)
self.y_train = y_train
self.y_val = y_valid
elif self.algorithm == "Linear Regression":
print("Linear Regression izabran")
# UBACI OVDE POZIV LINEAR REGRESSION
df = load_file(self.file_path)
new_df = preprocess(df)
x_train, y_train, x_valid, y_valid, self.train, self.valid = train_valid_split(new_df)
self.y_predict = linearregression.run_regression(x_train, y_train, x_valid, y_valid)
self.y_train = y_train
self.y_val = y_valid
plt.plot(self.y_train)
plt.plot(self.y_val)
plt.plot(self.y_predict)
plt.show()
elif self.algorithm == "Prophet":
print("Prophet")
df = load_file(self.file_path)
new_df = preprocess(df)
x_train, y_train, x_valid, y_valid, self.train, self.valid = train_valid_split(new_df)
self.y_predict = prophet_predict(self.train, self.valid)
self.y_train = y_train
self.y_val = y_valid
plt.plot(self.y_train)
plt.plot(self.y_val)
plt.plot(self.y_predict)
plt.show()
print(self.y_predict)
print("Zavrsio obucavanje i predikciju")
p1 = figure(x_axis_type="datetime", title="Stock Closing Prices")
p1.grid.grid_line_alpha = 0.3
p1.xaxis.axis_label = 'Date'
p1.yaxis.axis_label = 'Price'
plot_dates = df['Date']
print(self.valid)
plot_dates = plot_dates[-len(self.y_predict):]
p1.line(plot_dates, self.valid['Close'], color='#A6CEE3', legend=self.mode)
p1.line(plot_dates, self.y_predict, color='#B2DF8A', legend="Predicted "+self.mode)
output_file("stocks.html", title="Stocks prediction")
show(gridplot([[p1]], plot_width=500, plot_height=500))
def test_preprocess(self):
doc = 'Preprocess this string and output array of tokens'
assert preprocess(doc) == [
'preprocess', 'string', 'output', 'array', 'token'
]
from main import preprocess, extract_contours, cleanAndRead
windowname = "Live Video"
cap = cv2.VideoCapture(0)
t = 3
if cap.isOpened():
ret, frame = cap.read()
else:
ret = False
while ret:
ret, img = cap.read()
#time.sleep(t)
cv2.imshow(windowname, img)
print("DETECTING PLATE . . .")
#img = cv2.imread("testData/Final.JPG")
threshold_img = preprocess(img)
contours = extract_contours(threshold_img)
#if len(contours)!=0:
#print len(contours) #Test
# cv2.drawContours(img, contours, -1, (0,255,0), 1)
# cv2.imshow("Contours",img)
# cv2.waitKey(0)
cleanAndRead(img, contours)
if cv2.waitKey(1) == 27:
break
cv2.destroyAllWindows()
cap.release()
modelctrl = 1 #1 --> Ada, 0 --> SVM
###################
if datactrl == "PC1":
dataset_path = 'MDP csv/PC01.csv'
if datactrl == "PC2":
dataset_path = 'MDP csv/PC02.csv'
if datactrl == "PC3":
dataset_path = 'MDP csv/PC03.csv'
if datactrl == "PC4":
dataset_path = 'MDP csv/PC04.csv'
if datactrl == "PC5":
dataset_path = 'MDP csv/PC05.csv'
print("dataset_path-->", dataset_path)
return dataset_path
feature_data, target_data = preprocess(dataset_path, datactrl)
print('feature_data')
print(feature_data)
print(feature_data.shape)
print('target_data')
print(target_data)
print(target_data.shape)
discretize_data = discretize(feature_data)
print('\n')
print("*** Discretize Data ***")
print(discretize_data)
print(discretize_data.shape)
concat_data = 0