class Trader():
"""
Purpose is to trade live on the Binance Crypto Exchange
Attributes
-----------
symbol: str
client: Client object - From binance.py
capital: float - represents how much money is in account
leverage: float - how many multiples of your money you want to trade
tf: int - timeframe
df: pd.DataFrame - data that will be analyzed.
Methods
----------
load_account
set_leverage
set_timeframe
get_position
get_necessary_data
set_asset
make_row
load_existing_asset
start_trading
Please look at each method for descriptions
"""
def __init__(self, binance=True, qtrade=False, db=False, ib=False):
if binance:
self.binance = Client()
# self.ftx = FtxClient()
self.capital = None
self.leverage = 1
self.tf = None
self.executor = TradeExecutor()
self.loader = _DataLoader(db=db, qtrade=qtrade, ib=ib)
self.illustrator = Illustrator()
self.strategy = FabStrategy()
self.order_history = pd.DataFrame()
self.trade_replay = []
self.precisions = {
symbol_dic['symbol']: symbol_dic['quantityPrecision']
for symbol_dic in self.loader.binance.futures_exchange_info()
['symbols']
}
def load_account(self, additional_balance: int = 0) -> str:
"""
Sign in to account using API_KEY and using Binance API
"""
# local_path_binance = os.path.dirname(os.path.dirname(os.path.abspath("__file__"))) + r"\local\binance_api.txt"
# local_path_ftx = os.path.dirname(os.path.dirname(os.path.abspath("__file__"))) + r"\local\ftx_api.txt"
# info_binance = pd.read_csv(local_path_binance).set_index('Name')
# info_ftx = pd.read_csv(local_path_ftx).set_index('Name')
API_KEY_BINANCE = config('API_KEY_BINANCE')
API_SECRET_BINANCE = config('API_SECRET_BINANCE')
# API_KEY_FTX = config('API_KEY_FTX')
# API_SECRET_FTX = config('API_SECRET_FTX')
self.binance = Client(api_key=API_KEY_BINANCE,
api_secret=API_SECRET_BINANCE)
# self.ftx = FtxClient(api_key=API_KEY_FTX, api_secret=API_SECRET_FTX)
return "Connected to Binance"
def get_capital(self, additional_balance=4944.2649865):
initial_margin = float(
self.binance.futures_account()['totalInitialMargin'])
available_balance = float(
self.binance.futures_account()['availableBalance'])
self.capital = (initial_margin + available_balance +
additional_balance) * self.leverage
return self.capital
def load_futures_trading_history(self, start_time, end_time=None):
df = pd.DataFrame()
ranges = Helper.determine_timestamp_positions(start_time,
end_time,
limit=1)
for index, timestamp in enumerate(ranges):
try:
df = df.append(
pd.DataFrame(
self.binance.futures_account_trades(
startTime=ranges[index],
endTime=ranges[index + 1])))
except (IndexError) as e:
pass
return df.reset_index(drop=True)
def get_tickers(self):
tickers = pd.DataFrame(self.binance.futures_ticker())
tickers = tickers.rename(
{
'lastPrice': 'close',
'openPrice': 'open',
'highPrice': 'high',
'lowPrice': 'low',
'openTime': 'timestamp'
},
axis=1)
tickers['date'] = Helper.millisecond_timestamp_to_datetime(
tickers['timestamp'])
return tickers[[
'symbol', 'timestamp', 'date', 'open', 'high', 'low', 'close',
'volume'
]].set_index('symbol')
def show_live_chart(self, symbol, tf, refresh_rate=1):
df_tf_sma = self.show_current_chart(symbol, tf)
while True:
Helper.sleep(refresh_rate)
clear_output(wait=True)
df_tf_sma = self.show_current_chart(symbol, tf, data=df_tf_sma)
def show_current_chart(self, symbol=None, tf=None, data=()):
if len(data) == 0:
data = self.set_asset(symbol,
tf,
max_candles_needed=375,
futures=True)
else:
row = self.set_asset(symbol,
tf,
max_candles_needed=1,
futures=True)
if data['date'].iloc[-1] == row['date'].iloc[-1]:
data = data[:-1].append(row)
else:
data = data.append(row)
tf_data = data['tf'].iloc[0]
df_tf = self.loader._timeframe_setter(
data, tf // tf_data, keep_last_row=True) if tf != tf_data else data
return self.illustrator.graph_df(df_tf, sma=False)
def get_necessary_data(self,
symbol: str,
tf: int,
max_candles_needed: int = 235) -> pd.DataFrame:
df = pd.DataFrame()
start_datetime = str(
Helper.datetime_from_tf(tf,
daily_1m_candles=1440,
max_candles_needed=max_candles_needed)[0])
df = df.append(
self.loader.get_binance_candles(symbol,
tf,
start_date=start_datetime))
df['symbol'] = [symbol for _ in range(len(df))]
return df.drop_duplicates()
def get_necessary_data_futures(
self,
symbol: str,
tf: int,
max_candles_needed: int = 245) -> pd.DataFrame:
"""
Gets the minimum necessary data to trade this asset.
Note: This method is used as a way to tackle the 1000 candle limit that is currently on the Binance API.
A discrete set of ~1000 group candles will be determined, and then the data will be extracted from each,
using the _get_binance_futures_candles method, and then all of them will be merged together.
Parameters:
symbol: str - Symbol of price ticker Ex. "BTCUSDT", "ETHUSDT"
tf: int - Timeframe wanted Ex. 1, 3, 5, 77, 100
max_candles_needed: int - Maximum candles needed in the desired timeframe Ex. 231, 770, 1440
:return pd.DataFrame of candlestick data
"""
now = time.time()
df = pd.DataFrame()
ranges = Helper.determine_candle_positions(max_candles_needed, tf)
for i in range(len(ranges)):
try:
df = df.append(
self.loader._get_binance_futures_candles(
symbol, tf, int(ranges[i]), int(ranges[i + 1]), now))
except IndexError:
pass
df['symbol'] = [symbol for _ in range(len(df))]
return df.drop_duplicates()
def get_single_asset_signals(self,
screener,
symbol,
tf,
enter=True,
shift=0):
df = self.set_asset(symbol, tf, max_candles_needed=245)
binance_df = int(df['tf'].iloc[0])
df_tf = self.loader._timeframe_setter(df,
tf // binance_df,
keep_last_row=True,
shift=shift)
strat = FabStrategy()
df_ma = strat.load_data(df_tf)
return df_ma, screener._check_for_tf_signals(df_ma,
max_candle_history=10,
enter=enter)
def set_asset(self,
symbol: str,
tf: int,
max_candles_needed: int = 245,
futures=False) -> pd.DataFrame:
binance_tf = Helper.find_greatest_divisible_timeframe(tf)
new_max_candles = max_candles_needed * (tf // binance_tf)
if futures:
df = self.get_necessary_data_futures(symbol, binance_tf,
new_max_candles)
return df
df = self.get_necessary_data(symbol, binance_tf, new_max_candles)
self.df, self.tf, self.symbol = df, tf, symbol
return df
def get_current_trade_progress(self, key=()) -> pd.DataFrame:
if len(key) == 0:
key = self.key
df = pd.DataFrame(self.binance.futures_position_information())
df[['positionAmt', 'entryPrice',
'markPrice']] = df[['positionAmt', 'entryPrice',
'markPrice']].astype(float)
df = df[df['positionAmt'] != 0][[
'symbol', 'positionAmt', 'entryPrice', 'markPrice'
]].reset_index(drop=True)
df = df.rename(
{
'positionAmt': 'size',
'entryPrice': 'enter price',
'markPrice': 'current price'
},
axis=1)
df['side'] = np.where(df['size'] > 0, 'BUY', 'SELL')
df['size'] = df['size'].abs()
df['usd size'] = df['size'] * df['enter price']
short_profitability = (
(0.9996**2) *
(2 - df['current price'] / df['enter price']) - 1) * 100
long_profitability = (
(0.9996**2) * (df['current price'] / df['enter price']) - 1) * 100
df['pnl (%)'] = np.where(df['side'] == 'SELL', short_profitability,
long_profitability)
df['pnl (USD)'] = df['usd size'] * (df['pnl (%)'] / 100)
df['share'] = df['usd size'] / df['usd size'].sum()
df = df.round(2)
df = df[[
'symbol', 'side', 'enter price', 'size', 'usd size',
'current price', 'pnl (%)', 'pnl (USD)', 'share'
]]
current_positions = df
if type(key) != type(None):
recent_signals = pd.read_csv(
r"C:\Users\haseab\Desktop\Python\PycharmProjects\FAB\local\Workers\Recent signals.txt"
)
current_positions['tf'] = [
int(recent_signals.set_index('symbol').loc[symbol, 'tf'])
for symbol in current_positions['symbol'].values
]
current_positions = self.key[['win loss rate',
'avg pnl']].reset_index().merge(
current_positions,
on=['symbol', 'tf'])
self.current_positions = current_positions
return current_positions
def _check_trade_close(self, screener, current_positions):
trades_to_close = []
for symbol, tf in current_positions.reset_index().set_index(
['symbol', 'tf']).index:
try:
df_tf = screener.df_dic[(symbol, tf)]
df_ma = self.strategy.load_data(df_tf)
display(Markdown(f"## {symbol} {tf}"))
display(self.illustrator.graph_df(df_ma[-50:], flat=True))
enter_signal, enter_x_last_row, *rest = screener._check_for_tf_signals(
df_tf, max_candle_history=10, enter=True)
close_signal, exit_x_last_row, *rest = screener._check_for_tf_signals(
df_tf, max_candle_history=10, enter=False)
self.df_tf, self.close_signal, self.exit_x_last_row = df_tf, close_signal, exit_x_last_row,
if close_signal:
# raise Exception(close_signal, exit_x_last_row, enter_x_last_row)
if enter_x_last_row and exit_x_last_row < enter_x_last_row:
continue
last_entry_date = pd.Timestamp(
self.recent_signals.reset_index().set_index(
['symbol', 'tf']).loc[(symbol, tf), 'date'])
last_exit_date = self.df['date'].iloc[self.exit_x_last_row]
if last_entry_date > last_exit_date:
print('failed')
continue
trades_to_close.append(symbol)
print(f'{symbol} Trade Closed')
except AttributeError as e:
return []
return trades_to_close
def trade_free_capital(self, executor, current_positions,
remaining_to_invest, trades_to_enter):
print("-----------------------------------------------")
print(f"Free capital: {remaining_to_invest} USD")
print("-----------------------------------------------", end="\n\n\n")
final_trades_to_enter = self.check_against_max_trade_size(
trades_to_enter, current_positions, self.max_trade_size)
self.order_history = self.order_history.append(
executor.enter_market(client=self.binance,
symbol_side_rule=final_trades_to_enter,
remaining_capital=remaining_to_invest,
number_of_trades=len(trades_to_enter),
reason='free_capital',
max_trade_size=self.max_trade_size))
display(self.order_history.tail(len(trades_to_enter)))
self.order_history.to_csv("FAB Order History.csv")
self.trade_replay.append(
self.get_current_trade_progress(printout=False))
def calculate_remaining_capital(self, current_positions, capital):
total_invested = current_positions['usd size'].sum(
) if len(current_positions) > 0 else 0
remaining_to_invest = capital - total_invested
return remaining_to_invest
def get_positions_amount(self,
list_of_symbols: list,
full_close=False,
divide_by=1,
max_trades=3) -> list:
"""
Gets the total amount of current position for a given symbol
Parameters:
------------
symbol: str
Ex. "BTCUSDT"
"ETHUSDT"
Returns float
Ex. If a 1.000 BTC position is open, it will return 1.0
"""
# [{'symbol':'CHRUSDT', 'positionAmt':1}]
position_values = {}
for symbol in list_of_symbols:
last_price = float(
self.binance.futures_mark_price(symbol=symbol)['markPrice'])
for coin_futures_info in self.binance.futures_position_information(
):
if coin_futures_info['symbol'] == symbol:
amount = float(coin_futures_info["positionAmt"])
amount = round(amount, self.precisions[symbol])
if divide_by > 1:
amount = round(amount - amount / divide_by,
self.precisions[symbol])
position_values[
symbol] = amount if amount > 0 else 10**(
-self.precisions[symbol])
if full_close:
position_values[symbol] = amount
else:
return position_values
return position_values
def get_key(self,
essential_metrics,
win_loss_threshold=0.350,
pnl_threshold=1.025,
data_point_threshold=5):
key = essential_metrics[
essential_metrics['amount of data'] >= data_point_threshold]
key = key[key['avg pnl'] > pnl_threshold]
key = key[key['win loss rate'] > win_loss_threshold]
key['win loss rate'] = key['win loss rate'].round(1)
key['wnl drawdown'] = 0.4 * key['win loss rate'] + 0.6 * key[
'longest drawdown']
master_key = key.reset_index().set_index(['symbol', 'tf']).sort_values(
'wnl drawdown', ascending=False)
# master_key = key.reset_index().set_index(['symbol', 'tf']).sort_values(by=['win loss rate', 'avg pnl'], ascending=[False, False])
self.key = master_key
return master_key
def _profit_optimization(self, key, current_positions, trades_to_enter,
number_of_trades):
"""
Returns:
final_trades_to_enter: Ex. [('ZILUSDT', "Short", 'Rule 2'), ('BTCUSDT', 'Long', 'Rule 1'), ("ETHUSDT", "Short", 'Rule 3')]
trades_to_close: Ex. ['ADAUSDT']
"""
print()
print("Optimizing Trades:... ")
temp_metrics = key.reset_index().set_index(
['symbol', 'side', 'rule no'])
trades_of_interest = [(symbol,
int(temp_metrics.loc[(symbol, side, rule),
'tf']))
for symbol, side, rule in trades_to_enter
if (symbol, side, rule) in temp_metrics.index]
if not trades_of_interest:
print("Didn't make it past key filter", end='\n\n')
return [], []
current_position_symbols_tf_pair = Helper().nparray_to_tuple(
current_positions.reset_index()[['symbol', 'tf']].values)
all_symbols = list(
set(trades_of_interest).union(
set(current_position_symbols_tf_pair)))
self.current_position_symbols_tf_pair, self.trades_of_interest, self.all_symbols = current_position_symbols_tf_pair, trades_of_interest, all_symbols
self.top_x_symbols = sorted(
all_symbols,
key=lambda x: (key.loc[x, 'win loss rate'], key.loc[x, 'avg pnl']),
reverse=True)[:number_of_trades]
partial_trades_to_enter = list(
set(self.top_x_symbols).intersection(
set(trades_of_interest)).difference(
set(current_position_symbols_tf_pair)))
trades_to_close = list(
set(current_position_symbols_tf_pair).difference(
set(self.top_x_symbols)))
trades_to_close = [symbol for symbol, tf in trades_to_close]
final_trades_to_enter = [
(symbol, side, rule)
for (symbol,
tf), (symbol, side,
rule) in zip(trades_of_interest, trades_to_enter)
if (symbol, tf) in partial_trades_to_enter
]
for tup in trades_of_interest:
if tup in final_trades_to_enter:
raise Exception(tup, trades_of_interest, final_trades_to_enter)
return trades_to_close, final_trades_to_enter
def get_dividing_factor(self, current_positions, final_trades_to_enter,
number_of_trades):
dividing_factor = (len(current_positions) +
len(final_trades_to_enter)) / len(current_positions)
minimum_threshold = current_positions['usd size'].sum() * 0.95 // len(
current_positions)
if dividing_factor > 1 and minimum_threshold > (
self.capital) // number_of_trades:
print(minimum_threshold, (self.capital) // number_of_trades)
return dividing_factor
return 1
def optimize_trades(self, executor, current_positions, trades_to_enter,
number_of_trades):
if len(current_positions) != 0:
self.position_capital = self.get_capital()
trades_to_close, final_trades_to_enter = self._profit_optimization(
self.key, current_positions, trades_to_enter, number_of_trades)
if trades_to_close:
self.order_history = self.order_history.append(
executor.exit_market(self.binance,
self.get_positions_amount(
trades_to_close, full_close=True),
reason='lower rank'))
display(self.order_history)
current_positions = self.get_current_trade_progress(
printout=False)
if final_trades_to_enter:
dividing_factor = self.get_dividing_factor(
current_positions, final_trades_to_enter, number_of_trades)
positions_amount = self.get_positions_amount(
current_positions['symbol'].values,
divide_by=dividing_factor,
max_trades=number_of_trades)
if positions_amount:
self.order_history = self.order_history.append(
executor.exit_market(self.binance,
positions_amount,
reason='making space'))
self.position_capital = self.get_capital()
final_trades_to_enter = self.check_against_max_trade_size(
final_trades_to_enter, current_positions,
self.max_trade_size)
self.remaining_capital = self.calculate_remaining_capital(
current_positions, self.position_capital)
self.order_history = self.order_history.append(
executor.enter_market(self.binance,
final_trades_to_enter,
self.remaining_capital,
len(final_trades_to_enter),
reason='higher rank',
max_trade_size=self.max_trade_size))
display(self.order_history)
else:
print("No Trades made")
self.trades_to_close, self.final_trades_to_enter = trades_to_close, final_trades_to_enter
self.order_history.to_csv("FAB Order History.csv")
self.trade_replay.append(
self.get_current_trade_progress(printout=False))
def close_any_old_trades(self, screener, executor, current_positions):
if len(current_positions) == 0:
return False
trades_to_close = self._check_trade_close(screener, current_positions)
if trades_to_close:
self.order_history = self.order_history.append(
executor.exit_market(self.binance,
self.get_positions_amount(
trades_to_close, full_close=True),
reason='trade closed'))
def update_trades_to_enter(self, current_positions, trades_to_enter):
if len(trades_to_enter) == 0:
trades_to_enter = [(symbol, self.executor.dic[side], None)
for symbol, side in current_positions[
['symbol', 'side']].values]
return trades_to_enter
def check_against_max_trade_size(self, trades_to_enter, current_positions,
max_trade_size):
final_trades_to_enter = []
for symbol, side, rule in trades_to_enter:
if symbol in current_positions[
'symbol'].values and current_positions.set_index(
'symbol').loc[symbol,
'usd size'] >= max_trade_size * 0.99:
print('worked')
continue
final_trades_to_enter.append((symbol, side, rule))
return final_trades_to_enter
def monitor_fab(self,
screener,
df_metrics,
tfs,
number_of_trades=3,
leverage=1,
recency=-1,
additional_balance=0,
max_trade_size=None):
# self.order_history = pd.DataFrame()
self.load_account()
executor = self.executor
self.leverage = leverage
essential_metrics = Helper.drop_extra_delays(df_metrics,
screener.tf_delay_match)
self.starting_capital = self.get_capital(
additional_balance=additional_balance)
self.max_trade_size = max_trade_size
self.key = self.get_key(essential_metrics)
# self.key.to_csv ('Optimization Function Key.csv')
now = Helper.current_minute_datetime()
while True:
if datetime.now() >= now + pd.Timedelta(1, 'minute'):
clear_output(wait=True)
print("Current time: \t", datetime.now(), end='\n\n')
current_positions = self.get_current_trade_progress(
key=self.key)
if len(current_positions) > number_of_trades:
raise Exception(
f"You have more than {number_of_trades} active trades! Close first"
)
if current_positions['usd size'].sum(
) > self.starting_capital * 1.5:
raise Exception('We are too overexposed!',
current_positions['usd size'].sum())
self.close_any_old_trades(screener, executor,
current_positions)
remaining_capital = self.calculate_remaining_capital(
current_positions, self.capital)
## trades_to_enter is a list of lists (Ex. [('BANDUSDT', 'Short', 'Rule2'), ('BCHUSDT', 'Short', 'Rule 1')]
## df_recent_signals is a regular screener dataframe
trades_to_enter, recent_signals = screener.top_trades(
trader=self,
df_metrics=df_metrics,
tfs=tfs,
n=number_of_trades,
recency=recency)
self.trades_to_enter, self.recent_signals = trades_to_enter, recent_signals
print("Signals:")
display(recent_signals)
trades_left = number_of_trades - len(current_positions)
if remaining_capital > 100 and trades_left > 0 and trades_to_enter:
self.trade_free_capital(executor, current_positions,
remaining_capital, trades_to_enter)
elif remaining_capital > 100 and len(current_positions) != 0:
trades_to_enter = self.update_trades_to_enter(
current_positions=current_positions,
trades_to_enter=trades_to_enter)
self.trade_free_capital(executor, current_positions,
remaining_capital, trades_to_enter)
elif not trades_to_enter:
Helper.sleep(60)
now = Helper.current_minute_datetime()
continue
else:
print("Leverage: ", self.leverage)
print("Trades to Enter: ", trades_to_enter)
print("Max Trades: ", number_of_trades)
self.optimize_trades(executor, current_positions,
trades_to_enter, number_of_trades)
Helper.sleep(60)
now = Helper.current_minute_datetime()
print()
Helper.output_loading()
from binance.client import Client
import pandas as pd
from pprint import pprint
cred = pd.read_csv("credentials.csv", index_col=0).loc["binance", :]
client = Client(cred.api_key, cred.secret_key)
SYMBOL = "BTCUSDT"
ticker = client.futures_ticker()
print(ticker)
klines = client.futures_klines(symbol=SYMBOL,
interval=Client.KLINE_INTERVAL_1DAY,
startTime=1546473600000,
endTime=1609632000000)
# print(len(klines))
# pprint(klines)
# [[1609459200000, '28948.19', '29668.86', '28627.12', '29337.16', '210716.398', 1609545599999, '6157505024.08511', 1511793, '101247.902', '2960175587.62208', '0'], [1609545600000, '29337.15', '33480.00', '28958.24', '32199.91', '545541.080', 1609631999999, '17122938614.70610', 3514545, '273388.463', '8578964529.70894', '0'], [1609632000000, '32198.41', '34832.25', '32000.02', '33054.53', '487486.989', 1609718399999, '16389111411.52760', 3325307, '238761.657', '8029365512.72767', '0']]
fund_rate = client.futures_funding_rate(symbol=SYMBOL,
startTime=1568102400001,
endTime=1612469148000,
limit=1000)
#
print(len(fund_rate))
pprint(fund_rate)
# futures_klines랑 그냥 같은 데이터 같음
continous_fut = client.futures_continous_klines(
pair=SYMBOL,
import os
from dotenv import load_dotenv
from binance import *
from binance.client import Client
from binance.websockets import BinanceSocketManager
load_dotenv()
api_key = os.getenv('api_key')
api_secret = os.getenv('api_secret')
client = Client(api_key, api_secret)
bm = BinanceSocketManager(client)
r = client.futures_ticker(symbol='BTCUSDT')
print(r)