def draw_one_fitted_picture(data, name, epco_num, time):
Close = data['Close']
Open = data['Open']
max_error = 0.01
max_val = Close[1]
min_val = Close[1]
#adapt the max_error
for val in Close:
if val > max_val:
max_val = val
if val < min_val:
min_val = val
for val in Open:
if val > max_val:
max_val = val
if val < min_val:
min_val = val
adapted_error = max_error * (max_val - min_val) * (max_val - min_val)
#bottom-up with simple interpolation
segments = segment.topdownsegment(Close, fit.interpolate,
fit.sumsquared_error, adapted_error)
segments1 = segment.topdownsegment(Open, fit.interpolate,
fit.sumsquared_error, adapted_error)
draw_plot(data, name)
draw_segments(segments, segments1)
#save figure
file_name = r'../fitted_figures/' + name + '/' + name + '_' + str(
time) + 'Y_' + str(epco_num) + '.jpg'
plt.savefig(file_name, bbox_inches='tight')
#plt.show()
plt.close()
def printError(my_dpi, data):
"""
All data contanining the stock price info are retrieved from the database given the stock name
:param my_dpi: dpi screen
:param data: data to be plot
:param max_error: maximum error allowed
"""
try:
stockFile = []
try:
for eachLine in data:
splitLine = eachLine.split(',')
if len(splitLine) == 2:
if 'values' not in eachLine:
stockFile.append(eachLine)
except Exception as e:
print(str(e), 'failed to organize pulled data.')
except Exception as e:
print(str(e), 'failed to pull pricing data')
try:
date, closep_raw = np.loadtxt(
stockFile,
delimiter=',',
unpack=True,
converters={0: mdates.bytespdate2num('%Y%m%d')})
closep = closep_raw[::-1]
max_error = 0.1
while max_error < float(2 * max(closep)):
segments1 = segment.slidingwindowsegment(closep, fit.interpolate,
fit.sumsquared_error,
max_error)
segments2 = segment.slidingwindowsegment(closep, fit.regression,
fit.sumsquared_error,
max_error)
err1 = str(evaluate_MSE(closep, segments1))
err4 = str(evaluate_MSE(closep, segments2))
segments1 = segment.topdownsegment(closep, fit.interpolate,
fit.sumsquared_error, max_error)
segments2 = segment.topdownsegment(closep, fit.regression,
fit.sumsquared_error, max_error)
err2 = str(evaluate_MSE(closep, segments1))
err5 = str(evaluate_MSE(closep, segments2))
segments1 = segment.bottomupsegment(closep, fit.interpolate,
fit.sumsquared_error,
max_error)
segments2 = segment.bottomupsegment(closep, fit.regression,
fit.sumsquared_error,
max_error)
err3 = str(evaluate_MSE(closep, segments1))
err6 = str(evaluate_MSE(closep, segments2))
print(
str(max_error) + " " + err1 + " " + err2 + " " + err3 + " " +
err4 + " " + err5 + " " + err6)
max_error += 0.1
except e:
print("Error")
def model_run(name):
msft = db[name]
msft_cursor = msft.find({})
data = []
time = datetime.datetime(2020, 6, 15, 9, 30, 00)
for i in msft_cursor:
if i['time'] >= time:
data.append(float(i['price'].replace(',', '')))
data = np.array(data)
segments = segment.topdownsegment(data, fit.interpolate,
fit.sumsquared_error, max_error)
box_feature = build_box_feature(segments, data)
print(box_feature)
train_data, test_data = train_test_split(box_feature,
test_size=0.33,
shuffle=False)
feature_vector = feature_extract(train_data)
print(feature_vector)
hmm = GaussianHMM(n_components=4)
hmm.fit(feature_vector)
possibile_outcomes = compute_all_possible_outcomes(n_steps_frac_low,
n_steps_frac_high,
n_steps_frac_change)
predict = predict_close_price_for_days(500, name)
# for stock in stocks:
# msft = db[stock]
# msft_cursor = msft.find({})
#
# data = []
# time = datetime.datetime(2020, 6, 15, 9, 30, 00)
# for i in msft_cursor:
# if i['time'] >= time:
# data.append(float(i['price'].replace(',', '')))
#
# data = np.array(data)
# segments = segment.topdownsegment(data, fit.interpolate, fit.sumsquared_error, max_error)
# box_feature = build_box_feature(segments, data)
# print(box_feature)
# train_data, test_data = train_test_split(box_feature, test_size=0.33, shuffle=False)
# feature_vector = feature_extract(train_data)
# print(feature_vector)
#
# hmm = GaussianHMM(n_components=4)
# hmm.fit(feature_vector)
# possibile_outcomes = compute_all_possible_outcomes(n_steps_frac_low, n_steps_frac_high, n_steps_frac_change)
# predict = predict_close_price_for_days(500,stock)
def seg_extract(data_name):
client = MongoClient(port=27017)
print('Connect MongoDB Successful')
db = client.StockAnalyze
msft = db[data_name]
cursor = msft.find({})
data = []
for i in cursor:
data.append(float(i['price'].replace(',', '')))
data = np.array(data)
segments = segment.topdownsegment(data, fit.interpolate,
fit.sumsquared_error, max_error)
input = feature_extract(segments)
output = extract_new_y_feature(segments)
size = int(input.shape[0] * 0.95)
train_x = input[:size]
train_y = output[1:size + 1]
test_x = input[size:-1]
test_y = output[size + 1:]
return train_x, train_y, test_x, test_y
def draw_window_API(my_dpi, max_error, stockToFetch):
"""
All data contanining the stock price info are retrieved from the database given the stock name
:param my_dpi: dpi screen
:param data: data to be plot
:param max_error: maximum error allowed
"""
fig = plt.figure(figsize=(1000 / my_dpi, 700 / my_dpi),
dpi=96,
edgecolor='k',
facecolor='black')
fig.suptitle("PIECEWISE SEGMENTATION INTERPOLATION",
fontsize="15",
color="white",
fontweight='bold',
bbox={
'facecolor': 'red',
'alpha': 0.5,
'pad': 10
})
try:
print('Currently Pulling', stockToFetch)
urlToVisit = 'http://chartapi.finance.yahoo.com/instrument/1.0/' + stockToFetch + '/chartdata;type=quote;range=5y/csv'
stockFile = []
try:
sourceCode = urllib.request.urlopen(urlToVisit).read().decode()
splitSource = sourceCode.split('\n')
for eachLine in splitSource:
splitLine = eachLine.split(',')
if len(splitLine) == 6:
if 'values' not in eachLine:
stockFile.append(eachLine)
except Exception as e:
print(str(e), 'failed to organize pulled data.')
except Exception as e:
print(str(e), 'failed to pull pricing data')
try:
date, closep, highp, lowp, openp, volume = np.loadtxt(
stockFile,
delimiter=',',
unpack=True,
converters={0: mdates.bytespdate2num('%Y%m%d')})
SP = len(date)
# First subplot
ax1 = plt.subplot2grid((3, 3), (0, 0), colspan=3)
segments = segment.slidingwindowsegment(closep, fit.interpolate,
fit.sumsquared_error,
max_error)
draw_plot(closep, plt, ax1, "Sliding window with interpolation")
draw_segments(segments, 'red')
plt.ylabel('Stock Price')
plt.title("Sliding window", color='w')
# Second subplot
ax2 = plt.subplot2grid((3, 3), (1, 0), colspan=3)
segments = segment.topdownsegment(closep, fit.interpolate,
fit.sumsquared_error, max_error)
draw_plot(closep, plt, ax2, "Sliding window with interpolation")
draw_segments(segments, 'green')
plt.ylabel('Stock Price')
plt.title("Top down", color='w')
# Third subplot
ax3 = plt.subplot2grid((3, 3), (2, 0), colspan=3)
segments = segment.bottomupsegment(closep, fit.interpolate,
fit.sumsquared_error, max_error)
draw_plot(closep, plt, ax3, "Sliding window with interpolation")
draw_segments(segments, 'blue')
plt.ylabel('Stock Price')
plt.title("Bottom up", color='w')
plt.subplots_adjust(hspace=0.3)
plt.show()
except e:
print("Error")
def draw_window(my_dpi, data, max_error):
"""
All data contanining the stock price info are retrieved from the database given the stock name
:param my_dpi: dpi screen
:param data: data to be plot
:param max_error: maximum error allowed
"""
fig = plt.figure(figsize=(1000 / my_dpi, 700 / my_dpi),
dpi=96,
facecolor='black')
fig.suptitle("PIECEWISE SEGMENTATION INTERPOLATION",
fontsize="15",
color="white",
fontweight='bold',
bbox={
'facecolor': 'red',
'alpha': 0.5,
'pad': 10
})
try:
stockFile = []
try:
for eachLine in data:
splitLine = eachLine.split(',')
if len(splitLine) == 2:
if 'values' not in eachLine:
stockFile.append(eachLine)
except Exception as e:
print(str(e), 'failed to organize pulled data.')
except Exception as e:
print(str(e), 'failed to pull pricing data')
try:
date, closep_raw = np.loadtxt(
stockFile,
delimiter=',',
unpack=True,
converters={0: mdates.bytespdate2num('%Y%m%d')})
closep = closep_raw[::-1]
print(max(closep))
max_error = max(closep) * 2.5
# First subplot
ax1 = plt.subplot2grid((3, 3), (0, 0), colspan=3)
segments = segment.slidingwindowsegment(closep, fit.interpolate,
fit.sumsquared_error,
max_error)
draw_plot(closep, plt, ax1, "Sliding window with interpolation")
draw_segments(segments, 'red')
plt.ylabel('Stock Price')
plt.title("SLIDING WINDOW - ERROR " +
str(evaluate_global_error(closep, segments)),
color='Yellow',
fontweight='bold')
# Second subplot
ax2 = plt.subplot2grid((3, 3), (1, 0), colspan=3)
segments = segment.topdownsegment(closep, fit.interpolate,
fit.sumsquared_error, max_error)
draw_plot(closep, plt, ax2, "Sliding window with interpolation")
draw_segments(segments, 'green')
plt.ylabel('Stock Price')
plt.title("TOP DOWN - ERROR " +
str(evaluate_global_error(closep, segments)),
color='Yellow',
fontweight='bold')
# Third subplot
ax3 = plt.subplot2grid((3, 3), (2, 0), colspan=3)
segments = segment.bottomupsegment(closep, fit.interpolate,
fit.sumsquared_error, max_error)
segments = segment.bottomupsegment(closep, fit.interpolate,
fit.sumsquared_error, max_error)
draw_plot(closep, plt, ax3, "Sliding window with interpolation")
draw_segments(segments, 'blue')
plt.ylabel('Stock Price')
plt.title("BOTTOM UP - ERROR " +
str(evaluate_global_error(closep, segments)),
color='Yellow',
fontweight='bold')
plt.subplots_adjust(hspace=0.3)
plt.show()
except e:
print("Error")
if start_price < end_price:
tmp = 0
elif start_price == end_price:
tmp = 1
else:
tmp = 2
for i in range(start, end):
label.append(tmp)
label.append(tmp)
return label
def draw_plot(data, plot_title):
plot(range(len(data)), data, alpha=0.8, color='red')
title(plot_title)
xlabel("Time seres in 5 second skip(s)")
ylabel("Price($)")
xlim((0, len(data) - 1))
legend()
figure()
segments = segment.topdownsegment(data, fit.interpolate, fit.sumsquared_error,
max_error)
label = add_label(segments)
print(label)
draw_plot(data, 'top-down with simple interpolation')
draw_segments(segments, data)
savefig('status box.png')
show()
# sliding window with regression figure() segments = segment.slidingwindowsegment(data, fit.regression, fit.sumsquared_error, max_error) draw_plot(data, "Sliding window with regression") draw_segments(segments) # bottom-up with regression figure() segments = segment.bottomupsegment(data, fit.regression, fit.sumsquared_error, max_error) draw_plot(data, "Bottom-up with regression") draw_segments(segments) # top-down with regression figure() segments = segment.topdownsegment(data, fit.regression, fit.sumsquared_error, max_error) draw_plot(data, "Top-down with regression") draw_segments(segments) # sliding window with simple interpolation figure() segments = segment.slidingwindowsegment(data, fit.interpolate, fit.sumsquared_error, max_error) draw_plot(data, "Sliding window with simple interpolation") draw_segments(segments) # bottom-up with simple interpolation figure() segments = segment.bottomupsegment(data, fit.interpolate, fit.sumsquared_error, max_error) draw_plot(data, "Bottom-up with simple interpolation") draw_segments(segments)
my_dpi=96
#sliding window with regression
fig = plt.figure(num=None, figsize=(800/my_dpi, 800/my_dpi), dpi=80, facecolor='w', edgecolor='k')
segments = segment.slidingwindowsegment(closep, fit.regression, fit.sumsquared_error, max_error)
draw_plot(closep,plt,"Sliding window with regression")
draw_segments(segments)
#bottom-up with regression
segments = segment.bottomupsegment(closep, fit.regression, fit.sumsquared_error, max_error)
draw_plot(closep,plt,"Bottom-up with regression")
draw_segments(segments)
#top-down with regression
#figure()
segments = segment.topdownsegment(closep, fit.regression, fit.sumsquared_error, max_error)
draw_plot(closep,plt,"Top-down with regression")
draw_segments(segments)
#sliding window with simple interpolation
#figure()
segments = segment.slidingwindowsegment(closep, fit.interpolate, fit.sumsquared_error, max_error)
draw_plot(closep,plt,"Sliding window with simple interpolation")
draw_segments(segments)
#bottom-up with simple interpolation
#figure()
segments = segment.bottomupsegment(closep, fit.interpolate, fit.sumsquared_error, max_error)
draw_plot(closep,plt,"Bottom-up with simple interpolation")
def draw_window_API(my_dpi, max_error, stockToFetch):
"""
All data contanining the stock price info are retrieved from the database given the stock name
:param my_dpi: dpi screen
:param data: data to be plot
:param max_error: maximum error allowed
"""
fig = plt.figure(figsize=(1000/my_dpi, 700/my_dpi), dpi=96, edgecolor='k', facecolor='black')
fig.suptitle("PIECEWISE SEGMENTATION REGRESSION", fontsize="15", color="white", fontweight='bold', bbox={'facecolor':'red', 'alpha':0.5, 'pad':10})
try:
print('Currently Pulling',stockToFetch)
urlToVisit = 'http://chartapi.finance.yahoo.com/instrument/1.0/'+stockToFetch+'/chartdata;type=quote;range=5y/csv'
stockFile =[]
try:
sourceCode = urllib.request.urlopen(urlToVisit).read().decode()
splitSource = sourceCode.split('\n')
for eachLine in splitSource:
splitLine = eachLine.split(',')
if len(splitLine) == 6:
if 'values' not in eachLine:
stockFile.append(eachLine)
except Exception as e:
print(str(e), 'failed to organize pulled data.')
except Exception as e:
print(str(e), 'failed to pull pricing data')
try:
date, closep, highp, lowp, openp, volume = np.loadtxt(stockFile, delimiter=',', unpack=True,
converters={0: mdates.bytespdate2num('%Y%m%d')})
SP = len(date)
# First subplot
ax1 = plt.subplot2grid((3, 3), (0, 0), colspan=3)
segments = segment.slidingwindowsegment(closep, fit.regression, fit.sumsquared_error, max_error)
draw_plot(closep,plt,ax1,"Sliding window with regression")
draw_segments(segments,'red')
plt.ylabel('Stock Price')
plt.title("Sliding window", color='w')
# Second subplot
ax2 = plt.subplot2grid((3, 3), (1, 0), colspan=3)
segments = segment.topdownsegment(closep, fit.regression, fit.sumsquared_error, max_error)
draw_plot(closep, plt, ax2, "Sliding window with regression")
draw_segments(segments,'green')
plt.ylabel('Stock Price')
plt.title("Top down", color='w')
# Third subplot
ax3 = plt.subplot2grid((3, 3), (2, 0), colspan=3)
segments = segment.bottomupsegment(closep, fit.regression, fit.sumsquared_error, max_error)
draw_plot(closep, plt, ax3, "Sliding window with regression")
draw_segments(segments,'blue')
plt.ylabel('Stock Price')
plt.title("Bottom up", color='w')
plt.subplots_adjust(hspace=0.3)
plt.show()
except e:
print("Error")
def draw_window(my_dpi, data):
"""
All data contanining the stock price info are retrieved from the database given the stock name
:param my_dpi: dpi screen
:param data: data to be plot
:param max_error: maximum error allowed
"""
fig = plt.figure(figsize=(1000/my_dpi, 700/my_dpi), dpi=96, facecolor='black')
fig.suptitle("PIECEWISE SEGMENTATION REGRESSION", fontsize="15", color="white", fontweight='bold', bbox={'facecolor':'red', 'alpha':0.5, 'pad':10})
try:
stockFile = []
try:
for eachLine in data:
splitLine = eachLine.split(',')
if len(splitLine) == 2:
if 'values' not in eachLine:
stockFile.append(eachLine)
except Exception as e:
print(str(e), 'failed to organize pulled data.')
except Exception as e:
print(str(e), 'failed to pull pricing data')
try:
date, closep_raw = np.loadtxt(stockFile, delimiter=',', unpack=True,
converters={0: mdates.bytespdate2num('%Y%m%d')})
closep = closep_raw[::-1]
max_closep = max(closep)
if(max_closep > 2.0):
max_error = max(closep)*2.7;
else:
max_error = max(closep)/2.5;
print(max_error)
# First subplot
ax1 = plt.subplot2grid((3, 3), (0, 0), colspan=3)
segments = segment.slidingwindowsegment(closep, fit.regression, fit.sumsquared_error, max_error)
draw_plot(closep,plt,ax1,"Sliding window with regression")
draw_segments(segments,'red')
plt.ylabel('Stock Price')
plt.title("SLIDING WINDOW - ERROR "+str(evaluate_global_error(closep, segments)), color='Yellow', fontweight='bold')
# Second subplot
ax2 = plt.subplot2grid((3, 3), (1, 0), colspan=3)
segments = segment.topdownsegment(closep, fit.regression, fit.sumsquared_error, max_error)
draw_plot(closep, plt, ax2, "Sliding window with regression")
draw_segments(segments,'green')
plt.ylabel('Stock Price')
plt.title("TOP DOWN - ERROR "+str(evaluate_global_error(closep, segments)), color='Yellow', fontweight='bold')
# Third subplot
ax3 = plt.subplot2grid((3, 3), (2, 0), colspan=3)
segments = segment.bottomupsegment(closep, fit.regression, fit.sumsquared_error, max_error)
draw_plot(closep, plt, ax3, "Sliding window with regression")
draw_segments(segments,'blue')
plt.ylabel('Stock Price')
plt.title("BOTTOM UP - ERROR "+str(evaluate_global_error(closep, segments)), color='Yellow', fontweight='bold')
plt.subplots_adjust(hspace=0.3)
plt.show()
except e:
print("Error")
figure()
segments = segment.slidingwindowsegment(data, fit.regression,
fit.sumsquared_error_regr, max_error)
draw_plot(data, "Sliding window with regression")
draw_segments(segments)
#bottom-up with regression
figure()
segments = segment.bottomupsegment(data, fit.regression,
fit.sumsquared_error_regr, max_error)
draw_plot(data, "Bottom-up with regression")
draw_segments(segments)
#top-down with regression
figure()
segments = segment.topdownsegment(data, fit.regression,
fit.sumsquared_error_regr, max_error)
draw_plot(data, "Top-down with regression")
draw_segments(segments)
#sliding window with simple interpolation
figure()
segments = segment.slidingwindowsegment(data, fit.interpolate,
fit.sumsquared_error_int, max_error)
draw_plot(data, "Sliding window with simple interpolation")
draw_segments(segments)
#bottom-up with simple interpolation
figure()
segments = segment.bottomupsegment(data, fit.interpolate,
fit.sumsquared_error_int, max_error)
draw_plot(data, "Bottom-up with simple interpolation")
