def initialize(context):
# Portfolio assets list
with open('sectors.json', 'r') as f:
sectors = json.load(f)[global_vars.SECTOR_NAME]
sectors = [sector.replace("/", "_") for sector in sectors]
# stupid catalyst uses bittrex symbols for symbols
if "BCH_BTC" in sectors:
sectors.remove("BCH_BTC")
sectors.append("BCC_BTC")
# too new
for asset in global_vars.ignored_assets:
if asset in sectors:
sectors.remove(asset)
# uhh i think we need more than one asset
if len(sectors) < 2:
print("only 1 asset in bunch for sector: {}".format(global_vars.SECTOR_NAME))
exit()
context.assets = symbols(*sectors)
context.nassets = len(context.assets)
# Set the time window that will be used to compute expected return
# and asset correlations
context.window = 15
# Set the number of days between each portfolio rebalancing
context.rebalance_period = 5
context.i = 0
def universe(context, today):
json_symbols = get_exchange_symbols('poloniex')
poloniex_universe_df = pd.DataFrame.from_dict(
json_symbols).transpose().astype(str)
poloniex_universe_df['base_currency'] = poloniex_universe_df.apply(
lambda row: row.symbol.split('_')[1],
axis=1)
poloniex_universe_df['market_currency'] = poloniex_universe_df.apply(
lambda row: row.symbol.split('_')[0],
axis=1)
poloniex_universe_df = poloniex_universe_df[
poloniex_universe_df['base_currency'] == context.base_currency]
poloniex_universe_df = poloniex_universe_df[
poloniex_universe_df.symbol != 'gas_btc']
# Markets currently not working on Catalyst 0.3.1
# 2017-01-01
# poloniex_universe_df = poloniex_universe_df[poloniex_universe_df.symbol != 'bcn_btc']
# poloniex_universe_df = poloniex_universe_df[poloniex_universe_df.symbol != 'burst_btc']
# poloniex_universe_df = poloniex_universe_df[poloniex_universe_df.symbol != 'dgb_btc']
# poloniex_universe_df = poloniex_universe_df[poloniex_universe_df.symbol != 'doge_btc']
# poloniex_universe_df = poloniex_universe_df[poloniex_universe_df.symbol != 'emc2_btc']
# poloniex_universe_df = poloniex_universe_df[poloniex_universe_df.symbol != 'pink_btc']
# poloniex_universe_df = poloniex_universe_df[poloniex_universe_df.symbol != 'sc_btc']
print(poloniex_universe_df.head())
date = str(today).split(' ')[0]
poloniex_universe_df = poloniex_universe_df[
poloniex_universe_df.start_date < date]
context.coins = symbols(*poloniex_universe_df.symbol)
print(len(poloniex_universe_df))
return poloniex_universe_df.symbol.tolist()
def initialize(context):
# Lista de activos
context.assets = ["btc_usdt"]
context.symbols = symbols("btc_usdt")
# Definir carácteristicas
context.row_features = ["open", "high", "low", "close"]
context.features = ["open", "high", "low", "close"]
# Activar siguiente línea para inlcuir velas Heiken Ashi
context.include_ha = True
if context.include_ha:
context.features.append("HAopen")
context.features.append("HAhigh")
context.features.append("HAlow")
context.features.append("HAclose")
# Activar siguiente línea para incluir Distribución de Acumulación de William
context.include_wadl = False
if context.include_wadl:
context.features.append("WAD")
context.render = True
context.min_max_scaler = preprocessing.MinMaxScaler(feature_range=(-1, 1))
def initialize(context):
# Portfolio assets list
context.assets = symbols('USDT_BTC','USDT_ETH','USDT_LTC','USDT_DASH', 'USDT_XMR')
context.nassets = len(context.assets)
# Set the time window that will be used to compute expected return and asset correlations
context.window = 180
# Set the number of days between each portfolio rebalancing
context.rebalance_period = 30
context.i = 0
def initialize(context):
log.info('=== initializing algo')
context.assets = symbols('ada_usdt', 'algo_usdt', 'atom_usdt', 'bat_usdt',
'bch_usdt', 'btc_usdt', 'comp_usdt', 'dash_usdt',
'eos_usdt', 'eth_usdt', 'link_usdt', 'ltc_usdt',
'neo_usdt', 'trx_usdt', 'xlm_usdt', 'xmr_usdt',
'xrp_usdt', 'xtz_usdt', 'zil_usdt', 'zrx_usdt')
context.nassets = len(context.assets)
context.set_commission(maker=0.0004, taker=0.0004)
context.set_slippage(slippage=0.0001)
def initialize(context):
# Portfolio assets list
context.assets = symbols('btc_usdt', 'eth_usdt', 'ltc_usdt', 'dash_usdt',
'xmr_usdt')
context.nassets = len(context.assets)
# Set the time window that will be used to compute expected return
# and asset correlations
context.window = 180
# Set the number of days between each portfolio rebalancing
context.rebalance_period = 30
context.i = 0
def initialize(context):
# Portfolio assets list
context.assets = symbols('btc_usdt', 'eth_usdt', 'ltc_usdt', 'dash_usdt',
'xmr_usdt')
context.nassets = len(context.assets)
# Set the time window that will be used to compute expected return
# and asset correlations
context.window = 180
# Set the number of days between each portfolio rebalancing
context.rebalance_period = 30
context.i = 0
def initialize(context):
# Lista de activos
context.assets = ["btc_usdt"]
context.symbols = symbols("btc_usdt")
# Definir carácteristicas
context.row_features = ["open", "high", "low", "close", "volume"]
context.features = ["open", "high", "low", "close", "volume"]
context.render = True
context.bar_period = 30
context.lookback_w = 30
context.lookforward_w = 5
context.vol_freq = 10
def initialize(context):
# Lista de activos
context.assets = ["btc_usdt"]
context.symbols = symbols("btc_usdt")
# Definir carácteristicas
context.row_features = ["open", "high", "low", "close", "volume"]
context.features = ["open", "high", "low", "close", "volume"]
context.initial_value = context.portfolio.starting_cash
context.i = 0
context.bar_period = 15
context.random_state = 42
context.n_estimators = 1000 # 400
context.max_deep = 10
context.n_jobs = 1
context.lookback_w = 30
context.lookforward_w = 5
context.model_trained = False
context.training_data_size = 700000 # 700000
context.model_pickle_directory = "/home/enzo/SupervisedLearningIntraday/TrainedModels/"
if os.path.isfile(context.model_pickle_directory + "meta_model.pkl"):
context.model = joblib.load(context.model_pickle_directory +
"model.pkl")
context.meta_model = joblib.load(context.model_pickle_directory +
"meta_model.pkl")
context.model_trained = True
print("Modelo Pre-Entrenado")
context.lags = 5
context.invested = False
context.cur_prices = np.zeros(context.lags + 1)
context.cur_returns = np.zeros(context.lags)
# Advanced parameters
context.minRet = 0.01
context.cpus = multiprocessing.cpu_count()
context.vol_delta = pd.Timedelta(hours=1)
context.vb_delta = pd.Timedelta(hours=1)
context.vol_freq = 10 # 10
context.numThreads = 24
context.d = 0.25 # este es el factor fraccional de Fractional Differentieted Features
def universe(context, lookback_date, current_date):
# get all the pairs for the given exchange
json_symbols = get_exchange_symbols(context.exchange)
# convert into a DataFrame for easier processing
df = pd.DataFrame.from_dict(json_symbols).transpose().astype(str)
df['base_currency'] = df.apply(lambda row: row.symbol.split('_')[1],
axis=1)
df['market_currency'] = df.apply(lambda row: row.symbol.split('_')[0],
axis=1)
# Filter all the pairs to get only the ones for a given base_currency
df = df[df['base_currency'] == context.base_currency]
# Filter all pairs to ensure that pair existed in the current date range
df = df[df.start_date < lookback_date]
df = df[df.end_daily >= current_date]
context.coins = symbols(*df.symbol) # convert all the pairs to symbols
return df.symbol.tolist()
def universe(context, lookback_date, current_date):
json_symbols = get_exchange_symbols(
context.exchange) # get all the pairs for the exchange
universe_df = pd.DataFrame.from_dict(json_symbols).transpose().astype(
str) # convert into a dataframe
universe_df['base_currency'] = universe_df.apply(
lambda row: row.symbol.split('_')[1], axis=1)
universe_df['market_currency'] = universe_df.apply(
lambda row: row.symbol.split('_')[0], axis=1)
# Filter all the exchange pairs to only the ones for a give base currency
universe_df = universe_df[universe_df['base_currency'] ==
context.base_currency]
# Filter all the pairs to ensure that pair existed in the current date range
universe_df = universe_df[universe_df.start_date < lookback_date]
universe_df = universe_df[universe_df.end_daily >= current_date]
context.coins = symbols(
*universe_df.symbol) # convert all the pairs to symbols
return universe_df.symbol.tolist()
def initialize(context):
many_assets = []
with open('sectors.json', 'r') as f:
sectors = json.load(f)
for sector in sectors:
many_assets += sectors[sector]
a_many_assets = []
# find the ones taht dont go back far enough!
for asset in many_assets:
asset = asset.replace('/', '_')
if asset not in global_vars.ignored_assets:
a_many_assets.append(asset)
a_many_assets.remove("BCH_BTC")
a_many_assets.append("BCC_BTC")
context.assets = symbols(*a_many_assets)
def universe(context, lookback_date, current_date):
# get all the pairs for the given exchange
json_symbols = get_exchange_symbols(context.exchange)
# convert into a DataFrame for easier processing
df = pd.DataFrame.from_dict(json_symbols).transpose().astype(str)
df['base_currency'] = df.apply(lambda row: row.symbol.split('_')[1],
axis=1)
df['market_currency'] = df.apply(lambda row: row.symbol.split('_')[0],
axis=1)
# Filter all the pairs to get only the ones for a given base_currency
df = df[df['base_currency'] == context.base_currency]
# Filter all pairs to ensure that pair existed in the current date range
df = df[df.start_date < lookback_date]
df = df[df.end_daily >= current_date]
context.coins = symbols(*df.symbol) # convert all the pairs to symbols
return df.symbol.tolist()
def initialize(context):
context.TimeSpan = 15
context.RollingWindowSize = 50
# Lista de activos
context.assets = [
"btc_usdt", "eth_usdt", "ltc_usdt", "xmr_usdt", "str_usdt", "xrp_usdt"
]
context.symbols = symbols("btc_usdt", "eth_usdt", "ltc_usdt", "xmr_usdt",
"str_usdt", "xrp_usdt")
# Definir carácteristicas
context.row_features = ["open", "high", "low", "close"]
context.features = ["open", "high", "low", "close"]
# Activar siguiente línea para inlcuir velas Heiken Ashi
context.include_ha = True
if context.include_ha:
context.features.append("HAopen")
context.features.append("HAhigh")
context.features.append("HAlow")
context.features.append("HAclose")
context.f = len(context.features)
context.m = len(context.assets)
context.n = context.RollingWindowSize
context.initial_value = context.portfolio.starting_cash
# Hiper-parámetros de modelo
context.tau = 0.001
context.actor_learning_rate = 0.0001
context.critic_learning_rate = 0.001
context.minibatch_size = 64
context.max_episodes = 50
context.max_ep_steps = 1000
context.total_steps = context.max_episodes * context.max_ep_steps
context.buffer_size = context.total_steps / 50 # 1000000
context.gamma = 0.99
context.epsilon = 0.1
context.epsilon_target = 0.01
context.epsilon_decay = (context.epsilon_target / context.epsilon)**(
1 / (int(context.total_steps * 0.8)))
# Valor de portafolio de entrenamiento
context.init_train_portfolio = context.initial_value
context.portfolio_value_memory = []
context.portfolio_value_memory.append(context.init_train_portfolio)
context.train_invested_quantity = 0.0
context.train_cash = context.init_train_portfolio
context.invested_value = 0.0
context.portfolio_w_memory = []
context.init_portfolio_w = []
for i in range(len(context.assets) + 1):
context.init_portfolio_w.append(0.0)
context.assets_quantity_invested = []
for i in range(len(context.assets)):
context.assets_quantity_invested.append(0.0)
context.init_portfolio_w[0] = 1.0
context.portfolio_w_memory.append(context.init_portfolio_w)
context.transaction_cost = 0.002
context.open_trade = False
context.model_trained = True
context.model_created = False
context.model_loaded = False
context.saver = None
context.last_train_operation = 2
context.model_path = "/home/enzo/PycharmProjects/DDPGPorfolioOptimization/TrainedModels/model.ckpt"
context.i = 0
context.bar_period = 30
context.waiting_period = 0
context.wait = 0
context.current_op = 2
def initialize(context):
log.info('initializing algo')
context.assets = symbols('btc_usd')
