import numpy as np
import pandas as pd
import pytest
from freqtrade.data.dataprovider import DataProvider
from freqtrade.enums import CandleType
from freqtrade.resolvers.strategy_resolver import StrategyResolver
from freqtrade.strategy import merge_informative_pair, stoploss_from_absolute, stoploss_from_open
from tests.conftest import generate_test_data, get_patched_exchange
def test_merge_informative_pair():
data = generate_test_data("15m", 40)
informative = generate_test_data("1h", 40)
cols_inf = list(informative.columns)
result = merge_informative_pair(data, informative, "15m", "1h", ffill=True)
assert isinstance(result, pd.DataFrame)
assert list(informative.columns) == cols_inf
assert len(result) == len(data)
assert "date" in result.columns
assert result["date"].equals(data["date"])
assert "date_1h" in result.columns
assert "open" in result.columns
assert "open_1h" in result.columns
assert result["open"].equals(data["open"])
assert "close" in result.columns
assert "close_1h" in result.columns
assert result["close"].equals(data["close"])
assert "volume" in result.columns
assert "volume_1h" in result.columns
assert result["volume"].equals(data["volume"])
# First 3 rows are empty.
# Pre-fillup doesn't happen as there is no prior candlw in the informative dataframe
assert result.iloc[0]["date_1h"] is pd.NaT
assert result.iloc[1]["date_1h"] is pd.NaT
assert result.iloc[2]["date_1h"] is pd.NaT
# Next 4 rows contain the starting date (0:00)
assert result.iloc[3]["date_1h"] == result.iloc[0]["date"]
assert result.iloc[4]["date_1h"] == result.iloc[0]["date"]
assert result.iloc[5]["date_1h"] == result.iloc[0]["date"]
assert result.iloc[6]["date_1h"] == result.iloc[0]["date"]
# Next 4 rows contain the next Hourly date original date row 4
assert result.iloc[7]["date_1h"] == result.iloc[4]["date"]
assert result.iloc[8]["date_1h"] == result.iloc[4]["date"]
informative = generate_test_data("1h", 40)
result = merge_informative_pair(data, informative, "15m", "1h", ffill=False)
# First 3 rows are empty
assert result.iloc[0]["date_1h"] is pd.NaT
assert result.iloc[1]["date_1h"] is pd.NaT
assert result.iloc[2]["date_1h"] is pd.NaT
# Next 4 rows contain the starting date (0:00)
assert result.iloc[3]["date_1h"] == result.iloc[0]["date"]
assert result.iloc[4]["date_1h"] is pd.NaT
assert result.iloc[5]["date_1h"] is pd.NaT
assert result.iloc[6]["date_1h"] is pd.NaT
# Next 4 rows contain the next Hourly date original date row 4
assert result.iloc[7]["date_1h"] == result.iloc[4]["date"]
assert result.iloc[8]["date_1h"] is pd.NaT
def test_merge_informative_pair_weekly():
# Covers roughly 2 months - until 2023-01-10
data = generate_test_data("1h", 1040, "2022-11-28")
informative = generate_test_data("1w", 40, "2022-11-01")
informative["day"] = informative["date"].dt.day_name()
result = merge_informative_pair(data, informative, "1h", "1w", ffill=True)
assert isinstance(result, pd.DataFrame)
# 2022-12-24 is a Saturday
candle1 = result.loc[(result["date"] == "2022-12-24T22:00:00.000Z")]
assert candle1.iloc[0]["date"] == pd.Timestamp("2022-12-24T22:00:00.000Z")
assert candle1.iloc[0]["date_1w"] == pd.Timestamp("2022-12-12T00:00:00.000Z")
candle2 = result.loc[(result["date"] == "2022-12-24T23:00:00.000Z")]
assert candle2.iloc[0]["date"] == pd.Timestamp("2022-12-24T23:00:00.000Z")
assert candle2.iloc[0]["date_1w"] == pd.Timestamp("2022-12-12T00:00:00.000Z")
# 2022-12-25 is a Sunday
candle3 = result.loc[(result["date"] == "2022-12-25T22:00:00.000Z")]
assert candle3.iloc[0]["date"] == pd.Timestamp("2022-12-25T22:00:00.000Z")
# Still old candle
assert candle3.iloc[0]["date_1w"] == pd.Timestamp("2022-12-12T00:00:00.000Z")
candle4 = result.loc[(result["date"] == "2022-12-25T23:00:00.000Z")]
assert candle4.iloc[0]["date"] == pd.Timestamp("2022-12-25T23:00:00.000Z")
assert candle4.iloc[0]["date_1w"] == pd.Timestamp("2022-12-19T00:00:00.000Z")
def test_merge_informative_pair_monthly():
# Covers roughly 2 months - until 2023-01-10
data = generate_test_data("1h", 1040, "2022-11-28")
informative = generate_test_data("1M", 40, "2022-01-01")
result = merge_informative_pair(data, informative, "1h", "1M", ffill=True)
assert isinstance(result, pd.DataFrame)
candle1 = result.loc[(result["date"] == "2022-12-31T22:00:00.000Z")]
assert candle1.iloc[0]["date"] == pd.Timestamp("2022-12-31T22:00:00.000Z")
assert candle1.iloc[0]["date_1M"] == pd.Timestamp("2022-11-01T00:00:00.000Z")
candle2 = result.loc[(result["date"] == "2022-12-31T23:00:00.000Z")]
assert candle2.iloc[0]["date"] == pd.Timestamp("2022-12-31T23:00:00.000Z")
assert candle2.iloc[0]["date_1M"] == pd.Timestamp("2022-12-01T00:00:00.000Z")
# Candle is empty, as the start-date did fail.
candle3 = result.loc[(result["date"] == "2022-11-30T22:00:00.000Z")]
assert candle3.iloc[0]["date"] == pd.Timestamp("2022-11-30T22:00:00.000Z")
# Merged on prior month
assert candle3.iloc[0]["date_1M"] == pd.Timestamp("2022-10-01T00:00:00.000Z")
# First candle with 1M data merged.
candle4 = result.loc[(result["date"] == "2022-11-30T23:00:00.000Z")]
assert candle4.iloc[0]["date"] == pd.Timestamp("2022-11-30T23:00:00.000Z")
assert candle4.iloc[0]["date_1M"] == pd.Timestamp("2022-11-01T00:00:00.000Z")
# Very first candle in the result dataframe
# Merged the latest informative candle before the start-date
candle5 = result.iloc[0]
assert candle5["date"] == pd.Timestamp("2022-11-28T00:00:00.000Z")
assert candle5["date_1M"] == pd.Timestamp("2022-10-01T00:00:00.000Z")
def test_merge_informative_pair_no_overlap():
# Covers roughly a day
data = generate_test_data("1m", 1440, "2022-11-28")
# Data stops WAY before the main data starts
informative = generate_test_data("1h", 40, "2022-11-01")
result = merge_informative_pair(data, informative, "1m", "1h", ffill=True)
assert isinstance(result, pd.DataFrame)
assert len(result) == len(data)
assert "date" in result.columns
assert result["date"].equals(data["date"])
assert "date_1h" in result.columns
# If there's no overlap, forward filling should not fill anything
assert result["date_1h"].isnull().all()
def test_merge_informative_pair_same():
data = generate_test_data("15m", 40)
informative = generate_test_data("15m", 40)
result = merge_informative_pair(data, informative, "15m", "15m", ffill=True)
assert isinstance(result, pd.DataFrame)
assert len(result) == len(data)
assert "date" in result.columns
assert result["date"].equals(data["date"])
assert "date_15m" in result.columns
assert "open" in result.columns
assert "open_15m" in result.columns
assert result["open"].equals(data["open"])
assert "close" in result.columns
assert "close_15m" in result.columns
assert result["close"].equals(data["close"])
assert "volume" in result.columns
assert "volume_15m" in result.columns
assert result["volume"].equals(data["volume"])
# Dates match 1:1
assert result["date_15m"].equals(result["date"])
def test_merge_informative_pair_lower():
data = generate_test_data("1h", 40)
informative = generate_test_data("15m", 40)
with pytest.raises(ValueError, match=r"Tried to merge a faster timeframe .*"):
merge_informative_pair(data, informative, "1h", "15m", ffill=True)
def test_merge_informative_pair_empty():
data = generate_test_data("1h", 40)
informative = pd.DataFrame(columns=data.columns)
result = merge_informative_pair(data, informative, "1h", "2h", ffill=True)
assert result["date"].equals(data["date"])
assert list(result.columns) == [
"date",
"open",
"high",
"low",
"close",
"volume",
"date_2h",
"open_2h",
"high_2h",
"low_2h",
"close_2h",
"volume_2h",
]
# We merge an empty dataframe, so all values should be NaN
for col in ["date_2h", "open_2h", "high_2h", "low_2h", "close_2h", "volume_2h"]:
assert result[col].isnull().all()
def test_merge_informative_pair_suffix():
data = generate_test_data("15m", 20)
informative = generate_test_data("1h", 20)
result = merge_informative_pair(
data, informative, "15m", "1h", append_timeframe=False, suffix="suf"
)
assert "date" in result.columns
assert result["date"].equals(data["date"])
assert "date_suf" in result.columns
assert "open_suf" in result.columns
assert "open_1h" not in result.columns
assert list(result.columns) == [
"date",
"open",
"high",
"low",
"close",
"volume",
"date_suf",
"open_suf",
"high_suf",
"low_suf",
"close_suf",
"volume_suf",
]
def test_merge_informative_pair_suffix_append_timeframe():
data = generate_test_data("15m", 20)
informative = generate_test_data("1h", 20)
with pytest.raises(ValueError, match=r"You can not specify `append_timeframe` .*"):
merge_informative_pair(data, informative, "15m", "1h", suffix="suf")
@pytest.mark.parametrize(
"side,profitrange",
[
# profit range for long is [-1, inf] while for shorts is [-inf, 1]
("long", [-0.99, 2, 30]),
("short", [-2.0, 0.99, 30]),
],
)
def test_stoploss_from_open(side, profitrange):
open_price_ranges = [
[0.01, 1.00, 30],
[1, 100, 30],
[100, 10000, 30],
]
for open_range in open_price_ranges:
for open_price in np.linspace(*open_range):
for desired_stop in np.linspace(-0.50, 0.50, 30):
if side == "long":
# -1 is not a valid current_profit, should return 1
assert stoploss_from_open(desired_stop, -1) == 1
else:
# 1 is not a valid current_profit for shorts, should return 1
assert stoploss_from_open(desired_stop, 1, True) == 1
for current_profit in np.linspace(*profitrange):
if side == "long":
current_price = open_price * (1 + current_profit)
expected_stop_price = open_price * (1 + desired_stop)
stoploss = stoploss_from_open(desired_stop, current_profit)
stop_price = current_price * (1 - stoploss)
else:
current_price = open_price * (1 - current_profit)
expected_stop_price = open_price * (1 - desired_stop)
stoploss = stoploss_from_open(desired_stop, current_profit, True)
stop_price = current_price * (1 + stoploss)
assert stoploss >= 0
# Technically the formula can yield values greater than 1 for shorts
# even though it doesn't make sense because the position would be liquidated
if side == "long":
assert stoploss <= 1
# there is no correct answer if the expected stop price is above
# the current price
if (side == "long" and expected_stop_price > current_price) or (
side == "short" and expected_stop_price < current_price
):
assert stoploss == 0
else:
assert pytest.approx(stop_price) == expected_stop_price
@pytest.mark.parametrize(
"side,rel_stop,curr_profit,leverage,expected",
[
# profit range for long is [-1, inf] while for shorts is [-inf, 1]
("long", 0, -1, 1, 1),
("long", 0, 0.1, 1, 0.09090909),
("long", -0.1, 0.1, 1, 0.18181818),
("long", 0.1, 0.2, 1, 0.08333333),
("long", 0.1, 0.5, 1, 0.266666666),
("long", 0.1, 5, 1, 0.816666666), # 500% profit, set stoploss to 10% above open price
("long", 0, 5, 10, 3.3333333), # 500% profit, set stoploss break even
("long", 0.1, 5, 10, 3.26666666), # 500% profit, set stoploss to 10% above open price
("long", -0.1, 5, 10, 3.3999999), # 500% profit, set stoploss to 10% belowopen price
("short", 0, 0.1, 1, 0.1111111),
("short", -0.1, 0.1, 1, 0.2222222),
("short", 0.1, 0.2, 1, 0.125),
("short", 0.1, 1, 1, 1),
("short", -0.01, 5, 10, 10.01999999), # 500% profit at 10x
],
)
def test_stoploss_from_open_leverage(side, rel_stop, curr_profit, leverage, expected):
stoploss = stoploss_from_open(rel_stop, curr_profit, side == "short", leverage)
assert pytest.approx(stoploss) == expected
open_rate = 100
if stoploss != 1:
if side == "long":
current_rate = open_rate * (1 + curr_profit / leverage)
stop = current_rate * (1 - stoploss / leverage)
assert pytest.approx(stop) == open_rate * (1 + rel_stop / leverage)
else:
current_rate = open_rate * (1 - curr_profit / leverage)
stop = current_rate * (1 + stoploss / leverage)
assert pytest.approx(stop) == open_rate * (1 - rel_stop / leverage)
def test_stoploss_from_absolute():
assert pytest.approx(stoploss_from_absolute(90, 100)) == 1 - (90 / 100)
assert pytest.approx(stoploss_from_absolute(90, 100)) == 0.1
assert pytest.approx(stoploss_from_absolute(95, 100)) == 0.05
assert pytest.approx(stoploss_from_absolute(100, 100)) == 0
assert pytest.approx(stoploss_from_absolute(110, 100)) == 0
assert pytest.approx(stoploss_from_absolute(100, 0)) == 1
assert pytest.approx(stoploss_from_absolute(0, 100)) == 1
assert pytest.approx(stoploss_from_absolute(0, 100, False, leverage=5)) == 5
assert pytest.approx(stoploss_from_absolute(90, 100, True)) == 0
assert pytest.approx(stoploss_from_absolute(100, 100, True)) == 0
assert pytest.approx(stoploss_from_absolute(110, 100, True)) == -(1 - (110 / 100))
assert pytest.approx(stoploss_from_absolute(110, 100, True)) == 0.1
assert pytest.approx(stoploss_from_absolute(105, 100, True)) == 0.05
assert pytest.approx(stoploss_from_absolute(105, 100, True, 5)) == 0.05 * 5
assert pytest.approx(stoploss_from_absolute(100, 0, True)) == 1
assert pytest.approx(stoploss_from_absolute(0, 100, True)) == 0
assert pytest.approx(stoploss_from_absolute(100, 99, is_short=True)) == 0.01010101
assert pytest.approx(stoploss_from_absolute(100, 90, is_short=True)) == 0.1111111
assert pytest.approx(stoploss_from_absolute(100, 1, is_short=True)) == 99.0
assert pytest.approx(stoploss_from_absolute(100, 1, is_short=True, leverage=5)) == 495.0
assert pytest.approx(stoploss_from_absolute(100, 90, is_short=True, leverage=5)) == 0.55555555
@pytest.mark.parametrize("trading_mode", ["futures", "spot"])
def test_informative_decorator(mocker, default_conf_usdt, trading_mode):
candle_def = CandleType.get_default(trading_mode)
default_conf_usdt["candle_type_def"] = candle_def
test_data_5m = generate_test_data("5m", 40)
test_data_30m = generate_test_data("30m", 40)
test_data_1h = generate_test_data("1h", 40)
data = {
("XRP/USDT", "5m", candle_def): test_data_5m,
("XRP/USDT", "30m", candle_def): test_data_30m,
("XRP/USDT", "1h", candle_def): test_data_1h,
("XRP/BTC", "1h", candle_def): test_data_1h, # from {base}/BTC
("LTC/USDT", "5m", candle_def): test_data_5m,
("LTC/USDT", "30m", candle_def): test_data_30m,
("LTC/USDT", "1h", candle_def): test_data_1h,
("LTC/BTC", "1h", candle_def): test_data_1h, # from {base}/BTC
("NEO/USDT", "30m", candle_def): test_data_30m,
("NEO/USDT", "5m", CandleType.SPOT): test_data_5m, # Explicit request with '' as candletype
("NEO/USDT", "15m", candle_def): test_data_5m, # Explicit request with '' as candletype
("NEO/USDT", "1h", candle_def): test_data_1h,
("ETH/USDT", "1h", candle_def): test_data_1h,
("ETH/USDT", "30m", candle_def): test_data_30m,
("ETH/BTC", "1h", CandleType.SPOT): test_data_1h, # Explicitly selected as spot
}
default_conf_usdt["strategy"] = "InformativeDecoratorTest"
strategy = StrategyResolver.load_strategy(default_conf_usdt)
exchange = get_patched_exchange(mocker, default_conf_usdt)
default_conf_usdt["candle_type_def"] = candle_def
strategy.dp = DataProvider({}, exchange, None)
mocker.patch.object(
strategy.dp, "current_whitelist", return_value=["XRP/USDT", "LTC/USDT", "NEO/USDT"]
)
assert len(strategy._ft_informative) == 7 # Equal to number of decorators used
informative_pairs = [
("XRP/USDT", "1h", candle_def),
("XRP/BTC", "1h", candle_def),
("LTC/USDT", "1h", candle_def),
("LTC/BTC", "1h", candle_def),
("XRP/USDT", "30m", candle_def),
("LTC/USDT", "30m", candle_def),
("NEO/USDT", "1h", candle_def),
("NEO/USDT", "30m", candle_def),
("NEO/USDT", "5m", candle_def),
("NEO/USDT", "15m", candle_def),
("NEO/USDT", "2h", CandleType.FUTURES),
("ETH/BTC", "1h", CandleType.SPOT), # One candle remains as spot
("ETH/USDT", "30m", candle_def),
]
for inf_pair in informative_pairs:
assert inf_pair in strategy.gather_informative_pairs()
def test_historic_ohlcv(pair, timeframe, candle_type):
return data.get(
(pair, timeframe or strategy.timeframe, CandleType.from_string(candle_type)),
pd.DataFrame(),
).copy()
mocker.patch(
"freqtrade.data.dataprovider.DataProvider.historic_ohlcv", side_effect=test_historic_ohlcv
)
analyzed = strategy.advise_all_indicators(
{p: data[(p, strategy.timeframe, candle_def)] for p in ("XRP/USDT", "LTC/USDT")}
)
expected_columns = [
"rsi_1h",
"rsi_30m", # Stacked informative decorators
"neo_usdt_rsi_1h", # NEO 1h informative
"rsi_NEO_USDT_neo_usdt_NEO/USDT_30m", # Column formatting
"rsi_from_callable", # Custom column formatter
"eth_btc_rsi_1h", # Quote currency not matching stake currency
"rsi",
"rsi_less", # Non-informative columns
"rsi_5m", # Manual informative dataframe
]
for _, dataframe in analyzed.items():
for col in expected_columns:
assert col in dataframe.columns
# Test non-available pairs
del data[("ETH/BTC", "1h", CandleType.SPOT)]
with pytest.raises(
ValueError, match=r"Informative dataframe for \(ETH\/BTC, 1h, spot\) is empty.*"
):
strategy.advise_all_indicators(
{p: data[(p, strategy.timeframe, candle_def)] for p in ("XRP/USDT", "LTC/USDT")}
)