def main(argv):
"""
Main
"""
TOKEN = os.environ['telegram_token']
CHAT_ID = os.environ['telegram_chat_id']
html_template_path ="./src/assets/email/email_template.html"
hora_now = f'{datetime.datetime.now() - datetime.timedelta(hours=6):%Y-%m-%d %H:%M:%S}'
parser = argparse.ArgumentParser(description='Apollo V 0.1 Beta')
parser.add_argument('-o','--order', action='store_true',
help='Determine if you want to make an order')
args = parser.parse_args()
make_order = args.order or False
if make_order:
# Hacer decisón para la posición
# Pone orden a precio de mercado
new_order = Order()
new_order.make_market_order(units, inv_instrument)
# Poner stop loss de orden
new_order.set_stop_loss(stop_loss)
# Poner Take profit
new_order.set_take_profit(take_profit)
html_file, html_path = create_html([op_buy, op_sell], html_template_path)
image_file, image_name = from_html_to_jpg(html_path)
# send emails
send_email('Predicciones de USDJPY',
os.environ['email_from'],
os.environ['email_members'],
os.environ['email_pass'],
html_file)
# send telegram
bot = telegram_bot(TOKEN)
bot.send_message(CHAT_ID, f"Predicciones de la hora {hora_now}")
bot.send_photo(CHAT_ID, f'{image_name}')
def main(argv):
"""
Main
"""
TOKEN = os.environ['telegram_token']
CHAT_ID = os.environ['telegram_chat_id']
html_template_path = "./src/assets/email/email_template.html"
tz_MX = pytz.timezone('America/Mexico_City')
datetime_MX = dt.now(tz_MX)
hora_now = f'{datetime_MX.strftime("%H:%M:%S")}'
parser = argparse.ArgumentParser(description='Apollo V 0.1 Beta')
parser.add_argument('-o',
'--order',
action='store_true',
help='Determine if you want to make an order')
parser.add_argument('-t',
'--time',
action='store_true',
help='Make order only if market is open')
parser.add_argument('-d',
'--debug',
action='store_true',
help='Debug mode on')
args = parser.parse_args()
make_order = args.order or False
market_sensitive = args.time or False
debug_mode = args.debug or False
logging.info(f'\nMarket sensitive: {market_sensitive}')
if market_sensitive and not market_open():
logging.info('Market Closed')
return
# Hacer decisón para la posición
decision = Decide(op_buy,
op_sell,
100000,
direction=0,
pips=1,
take_profit=0,
stop_loss=0)
decision.get_all_pips()
units = decision.pips * decision.direction * 1000
inv_instrument = 'USD_JPY'
take_profit = decision.take_profit
op_buy_new = decision.data_buy
op_sell_new = decision.data_sell
logging.info(f'\n{decision.decision}')
# Pone orden a precio de mercado
logging.info(
f'Units: {units}, inv_instrument: {inv_instrument} , take_profit: {take_profit}\n'
)
if make_order and units != 0:
new_order = Order(inv_instrument, take_profit)
new_order.make_market_order(units)
print(f'\nPrevious High Ask:{previous_high_ask}')
print(op_buy_new)
print(f'\nPrevious Low Bid: {previous_low_bid}')
print(op_sell_new)
# send telegram
if not debug_mode:
_, html_path = create_html(
[op_buy, op_sell, previous_high_ask, previous_low_bid],
html_template_path)
_, image_name = from_html_to_jpg(html_path)
logging.info('Se mandan predicciones a Telegram')
bot = telegram_bot(TOKEN)
if not make_order:
bot.send_message(CHAT_ID, f"TEST!!!!!")
bot.send_message(CHAT_ID, f"Predictions for the hour: {hora_now}")
bot.send_photo(CHAT_ID, image_name)
bot.send_message(CHAT_ID,
f"Best course of action: {decision.decision}")
def main(argv):
"""
Main
"""
TOKEN = os.environ['telegram_token']
CHAT_ID = os.environ['telegram_chat_id']
initial_pip = float(os.environ['initial_pip'])
html_template_path = "./assets/email/email_template.html"
tz_MX = pytz.timezone('America/Mexico_City')
datetime_MX = dt.now(tz_MX)
hora_now = f'{datetime_MX.strftime("%H:%M:%S")}'
parser = argparse.ArgumentParser(description='Apollo V 0.1 Beta')
parser.add_argument('-o',
'--order',
action='store_true',
help='Determine if you want to make an order')
parser.add_argument('-t',
'--time',
action='store_true',
help='Make order only if market is open')
parser.add_argument('-m',
'--model-version',
help='Model version folder simple name')
parser.add_argument('-i', '--instrument', help='instrument to trade')
parser.add_argument('-a', '--account', help='suffix of account to trade')
parser.add_argument('-s',
'--save',
action='store_true',
help='saves the predictions made')
args = parser.parse_args()
make_order = args.order or False
market_sensitive = args.time or False
model_version = args.model_version
instrument = args.instrument
account = args.account
save_preds = args.save or False
trading = Trading(model_version, instrument)
op_buy, op_sell, original_dataset = trading.predict()
previous_low_bid = str(
original_dataset[2]['USD_JPY_lowBid'].iloc[-2].round(3))
previous_high_ask = str(
original_dataset[2]['USD_JPY_highAsk'].iloc[-2].round(3))
logging.info(f'\nMarket sensitive: {market_sensitive}')
if market_sensitive and not market_open():
logging.info('Market Closed')
open_trades = openTrades(account)
current_pips = open_trades.get_pips_traded()
current_trades = open_trades.get_all_trades()
check_stop_loss(current_trades, account)
return
# Hacer decisón para la posición
decision = Decide(op_buy,
op_sell,
100000,
direction=0,
pips=initial_pip,
take_profit=0)
decision.get_all_pips()
units = decision.pips * decision.direction * 1000
# máximo de unidades en riesgo al mismo tiempo
pip_limit = float(os.environ['pip_limit'])
open_trades = openTrades(account)
current_pips = open_trades.get_pips_traded()
current_trades = open_trades.get_all_trades()
logging.info(f'Current units: {current_pips}')
logging.info(f'Max units: {pip_limit}')
logging.info(f'Units: {units}')
# si queremos hacer una operación (units puede ser positivo o negativo)
if units != 0:
if current_pips < pip_limit: # vemos si aún podemos hacer operaciones
# escogemos lo que podamos operar sin pasarnos del límite.
# el mínimo entre la unidades solicitadas o las disponibles
units = min(abs(units), pip_limit - current_pips) * \
decision.direction
if units == 0.0: # si encontramos que ya no hay
decision.decision += '\n*Units limit exceeded. Order not placed.'
else: # si ya hemos excedido operaciones
units = 0.0
decision.decision += '\n*Units limit exceeded. Order not placed.'
inv_instrument = 'USD_JPY'
take_profit = decision.take_profit
op_buy_new = decision.data_buy
op_sell_new = decision.data_sell
logging.info(f'inv_instrument: {inv_instrument}')
logging.info(f'take_profit: {take_profit}')
logging.info(f'\n{decision.decision}')
# Pone orden a precio de mercado
logging.info(
f'Units: {units}, inv_instrument: {inv_instrument} , take_profit: {take_profit}\n'
)
if make_order and units != 0:
new_trade = Trade(inv_instrument, units, take_profit=take_profit)
new_order = Order(new_trade, account)
new_order.make_market_order()
previous_low_bid += f' ({new_order.bid_price})'
previous_high_ask += f' ({new_order.ask_price})'
end = timer()
speed_time = end - start
logging.info('Apollo time to market: ' + str(end - start))
if save_preds:
logging.info(
'\n\n************* Saving predictions in Data Base **************'
)
save_order(account, model_version, inv_instrument, new_order,
decision.probability)
logging.info(
f'\n\n************* Saving dataset in Data Base **************'
)
save_input(account,
model_version,
hora_now,
inv_instrument,
original_dataset,
order_id=new_order.i_d)
else:
end = timer()
speed_time = end - start
logging.info(f'Apollo prediction time: {str(speed_time)} s')
logging.info(
f'\n\n************* Saving dataset in Data Base **************')
save_input(account, model_version, hora_now, inv_instrument,
original_dataset)
logging.info('\n\n ************* Checando trades activos **************')
check_stop_loss(current_trades, account)
logging.info(f'\nPrevious High Ask:{previous_high_ask}')
logging.info(op_buy_new)
logging.info(f'\nPrevious Low Bid: {previous_low_bid}')
logging.info(op_sell_new)
# send telegram
_, html_path = create_html([
op_buy, op_sell, previous_high_ask, previous_low_bid,
f'{model_version}'
], html_template_path)
_, image_name = from_html_to_jpg(html_path)
logging.info('Sending predictions to Telegram')
bot = telegram_bot(TOKEN)
bot.send_message(
CHAT_ID, f"Predictions for the hour: {hora_now} ({model_version})")
bot.send_photo(CHAT_ID, image_name)
bot.send_message(
CHAT_ID,
f"Best course of action ({model_version}): {decision.decision}\nApollo speed:{str(round(speed_time,3))}s"
)
logging.info(f'Apollo prediction time: {str(speed_time)} s')