def api_chicago():
months = json.loads(request.args.get('months'))
months = list(map(int, months))
days = json.loads(request.args.get('days'))
days = list(map(int, days))
# Checking for months and days filters and return error response if not set
if not months or not days:
return "Months and Days are required", 400
else:
return jsonify(fn.execute_analysis(fn.get_df('chicago', months, days)))
def api_chicago_users():
months = json.loads(request.args.get('months'))
months = list(map(int, months))
days = json.loads(request.args.get('days'))
days = list(map(int, days))
# Start and End slicing args of the get get_users function
start = int(json.loads(request.args.get('start')))
end = int(json.loads(request.args.get('end')))
df = fn.get_df('chicago', months, days)
users = fn.get_users(df, start, end).to_json(orient='index')
# Parsing users data as json object
parsed_users = json.loads(users)
return parsed_users
def update_graph(symbol,
type,
start_date,
end_date,
dict_tickers=dict_tickers):
url = 'https://www.alphavantage.co/query'
function = 'TIME_SERIES_DAILY'
outputsize = 'full'
response = get_response(url, function, outputsize, symbol)
df_response = get_df(response)
df_response = date_filter_df(df_response, start_date, end_date)
if type == 'ts':
if symbol in dict_tickers:
fig = plot_df_py(df_response, dict_tickers[symbol])
else:
fig = plot_df_py(df_response, symbol)
else:
if symbol in dict_tickers:
fig = plot_candlestick(df_response, dict_tickers[symbol])
else:
fig = plot_candlestick(df_response, symbol)
return fig
# -- ---------------------------------------------------------------------------------------------------------------- #
'''--------------------------------------------------------------
Crear modelo con features simbolicos
'''
# Generacion de un features simbolicas, agregadas al modelo
symbolic = ft.symbolic_features(p_x=features_divisa.iloc[:, 1:], p_y=features_divisa.iloc[:, 0])
nuevos_features = pd.DataFrame(symbolic['fit'], index=features_divisa.index)
# modelo
lm_model_s = ft.mult_reg(p_x=nuevos_features[:'01-01-2019'],
p_y=features_divisa.iloc[:, 0][:'01-01-2019'])
prediccion = ft.recursivo(nuevos_features, features_divisa, lm_model_s["ridge"]["model"]) #reales y pronostico
# -- ---------------------------------------------------------------------------------------------------------------- #
'''--------------------------------------------------------------
Backtest
'''
backtest = ft.backtest(prediccion, datos_divisa)
# -- ---------------------------------------------------------------------------------------------------------------- #
'''--------------------------------------------------------------
Metricas de atribucion al desempeño
'''
residuos = ft.get_residuos(backtest)
hetero = ft.check_hetero(residuos)
df = ft.get_df(backtest)
mad, lista = ft.f_estadisticas_mad(df, True)
'month',
MONTHS,
return_multi=True)
days = fnc.get_user_input(
'Which week day?\nPlease type out the full day name (e.g. Monday, Tuesday, ..) or type "all" to get all days.\n',
'week day',
DAYS,
return_multi=True)
print('\n' + '#' * 50)
print('Calculating the statistics of the selected dataset ...')
print('#' * 50 + '\n')
start_time = time.time()
result_df = fn.get_df(city[0], months, days)
result = fn.execute_analysis(result_df)
month = fnc.get_key(result[0]['month'], MONTHS).capitalize()
day = fnc.get_key(result[0]['day'], DAYS).capitalize()
hour = result[0]['hour']
start_station = result[1]['start']
end_station = result[1]['end']
trip = result[1]['trip']
total_time = result[2]['total']
mean_time = result[2]['mean']
subscribers = result[3]['subscriber']
customers = result[3]['customer']
males = result[3]['male']
females = result[3]['female']
first_birth = result[3]['min']
import numpy as np
table = pd.read_html('https://en.wikipedia.org/wiki/List_of_S%26P_500_companies')
sandp_raw = df = table[0]
sandp_symbols = sandp_raw["Symbol"]
for i in range(1, 10):
symbol = f.convert_to_string(sandp_symbols.iloc[i])
ticker = yf.Ticker(symbol)
name = ticker.info["longName"]
# contains total assets and total liabilities
df_balance_sheet = f.get_df(symbol, 'balance-sheet')
# contains Gross Profits
df_financials = f.get_df(symbol, 'financials')
# getting company stock price and eps
gross_profit_df = df_financials["Gross Profit"].head(1)
total_assets_df = df_balance_sheet["Total Assets"].head(1)
total_liabilities_df = df_balance_sheet["Total Liabilities Net Minority Interest"].head(1)
gross_profit_as_xml = gross_profit_df.iloc[0]
total_assets_as_xml = total_assets_df.iloc[0]
total_liabilities_as_xml = total_liabilities_df.iloc[0]
gross_profit = f.convert_to_float([str(s) for s in gross_profit_as_xml]) * 1000
total_assets = f.convert_to_float([str(s) for s in total_assets_as_xml]) * 1000
total_liabilities = f.convert_to_float([str(s) for s in total_liabilities_as_xml]) * 1000
from functions import get_response, get_df, plot_df, plot_df_py url = 'https://www.alphavantage.co/query' function = 'TIME_SERIES_DAILY' symbol = 'BA' outputsize = 'full' response = get_response(url, function, outputsize, symbol) df_response = get_df(response) # Plot with matplotlib plot_df(df_response, symbol)
ft.configuracion()
good_connect = False
key_file = st.sidebar.file_uploader("Clave:")
actualizar = st.sidebar.button("Actualizar Outcomes")
if actualizar:
if os.path.isfile("datos_pick"):
os.remove('datos_pick')
if key_file is not None:
if not os.path.isfile("datos_pick"):
key_json = ft.cargar_creadenciales(key_file)
cliente = ft.connect_to_sheet(key_json)
worksheet = ft.open_sheet(cliente, sheet_name)
df = ft.get_df(worksheet)
df.to_pickle('datos_pick')
else:
df = pd.read_pickle('datos_pick')
else:
st.warning("El archivo con los datos que has cargador no es correcto.")
st.stop()
menu = st.sidebar.selectbox(
'Menu:', options=['General', 'Estados de Contratacion', 'Trabajando'])
filtros = ft.opciones_filtros(df)
df = ft.filtrar(df, filtros)