def display_history_graph(self, historyDict, numberOfEpochs):
print(numberOfEpochs)
if numberOfEpochs is not None:
xAxis = list(range(0, numberOfEpochs))
core.add_line_series(self.plotName, "Dokladnosc", xAxis,
historyDict.history['accuracy'])
print(historyDict.history['accuracy'])
core.add_line_series(self.plotName, "Strata", xAxis,
historyDict.history['loss'])
print(historyDict.history['loss'])
def display_history_graph(self, historyDict, numberOfEpochs):
with simple.window(self.learningGraph, width=300, height=300):
core.add_separator()
core.add_plot(self.historyPlotName)
xAxis = range(0, numberOfEpochs)
core.add_line_series(self.historyPlotName, "Dokładnosc", xAxis,
historyDict['accuracy'])
core.add_line_series(self.historyPlotName, "Strata", xAxis,
historyDict['loss'])
def create(sender, data):
core.set_value("company_id",
"{} Model Created".format(model_company.split('_')[0]))
core.clear_plot("Pred")
predict, original = data
core.add_line_series("Pred",
"Prediction",
predict.index.tolist(),
predict[0].tolist(),
color=[255, 50, 50, 100])
def create_model(sender, data):
predict, original = model.create_model(model_company)
print(predict.index)
print(predict[0])
print(type(predict.index))
print(type(predict[0]))
core.add_line_series("Pred",
"Prediction",
predict.index.tolist(),
predict[0].tolist(),
color=[255, 50, 50, 100])
def __init__(self, label):
self.data = []
self.labels = []
self.plot_label = "##" + label
self.chart = lambda: core.add_line_series(
self.plot_label,
label,
y=self.data,
x=self.labels,
weight=2,
color=[0, 0, 255, 100],
)
def plot_callback(sender, data):
if data == "clear":
selected_companies.clear()
elif data != None:
if data not in selected_companies:
selected_companies.append(data)
else:
selected_companies.remove(data)
core.clear_plot("Plot")
cap_callback(None, selected_cap)
for i in range(0, len(selected_companies)):
company = selected_companies[i]
stocks = pd.read_csv("dataset/" + company)
data_x = list(range(1, len(stocks.index) + 1))
data_y = stocks['High'].tolist()
core.add_line_series("Plot",
company.split('_')[0],
data_x,
data_y,
color=colors[i % len(colors)])
def cap_callback(sender, data):
global selected_cap
selected_cap = data
core.clear_plot("Cap")
for i in range(0, len(selected_companies)):
company = selected_companies[i]
stocks = pd.read_csv("dataset/" + company, parse_dates=['Date'])
stocks['Date'] = pd.to_datetime(stocks['Date'],
unit='D',
errors='coerce')
if data == "daily":
data_x = list(range(1, len(stocks.index) + 1))
data_y = (stocks['High'] * stocks['Volume']).tolist()
core.add_line_series("Cap",
company.split('_')[0],
data_x,
data_y,
color=colors[i % len(colors)])
core.add_line_series("Cap",
"large-cap", [data_x[0], data_x[-1]],
[10000000000, 10000000000],
weight=3,
color=[255, 50, 50, 100])
core.add_line_series("Cap",
"mid-cap", [data_x[0], data_x[-1]],
[2000000000, 2000000000],
weight=3,
color=[200, 50, 50, 100])
core.add_line_series("Cap",
"small-cap", [data_x[0], data_x[-1]],
[300000000, 300000000],
weight=3,
color=[150, 50, 50, 100])
elif data == "monthly":
monthly_stocks = stocks.groupby(pd.Grouper(key="Date",
freq='1M')).mean()
data_x = list(range(1, len(monthly_stocks.index) + 1))
data_y = (monthly_stocks['High'] *
monthly_stocks['Volume']).tolist()
core.add_line_series("Cap",
company.split('_')[0],
data_x,
data_y,
color=colors[i % len(colors)])
core.add_line_series("Cap",
"large-cap", [data_x[0], data_x[-1]],
[10000000000, 10000000000],
weight=3,
color=[255, 50, 50, 100])
core.add_line_series("Cap",
"mid-cap", [data_x[0], data_x[-1]],
[2000000000, 2000000000],
weight=3,
color=[200, 50, 50, 100])
core.add_line_series("Cap",
"small-cap", [data_x[0], data_x[-1]],
[300000000, 300000000],
weight=3,
color=[150, 50, 50, 100])
elif data == "quarterly":
quarterly_stocks = stocks.groupby(pd.Grouper(key="Date",
freq='3M')).mean()
data_x = list(range(1, len(quarterly_stocks.index) + 1))
data_y = (quarterly_stocks['High'] *
quarterly_stocks['Volume']).tolist()
core.add_line_series("Cap",
company.split('_')[0],
data_x,
data_y,
color=colors[i % len(colors)])
core.add_line_series("Cap",
"large-cap", [data_x[0], data_x[-1]],
[10000000000, 10000000000],
weight=3,
color=[255, 50, 50, 100])
core.add_line_series("Cap",
"mid-cap", [data_x[0], data_x[-1]],
[2000000000, 2000000000],
weight=3,
color=[200, 50, 50, 100])
core.add_line_series("Cap",
"small-cap", [data_x[0], data_x[-1]],
[300000000, 300000000],
weight=3,
color=[150, 50, 50, 100])
elif data == "yearly":
yearly_stocks = stocks.groupby(pd.Grouper(key="Date",
freq='1Y')).mean()
data_x = list(range(1, len(yearly_stocks.index) + 1))
data_y = (yearly_stocks['High'] * yearly_stocks['Volume']).tolist()
core.add_line_series("Cap",
company.split('_')[0],
data_x,
data_y,
color=colors[i % len(colors)])
core.add_line_series("Cap",
"large-cap", [data_x[0], data_x[-1]],
[10000000000, 10000000000],
weight=3,
color=[255, 50, 50, 100])
core.add_line_series("Cap",
"mid-cap", [data_x[0], data_x[-1]],
[2000000000, 2000000000],
weight=3,
color=[200, 50, 50, 100])
core.add_line_series("Cap",
"small-cap", [data_x[0], data_x[-1]],
[300000000, 300000000],
weight=3,
color=[150, 50, 50, 100])