overfitting 0.4.0__tar.gz → 0.4.1__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (26) hide show
  1. {overfitting-0.4.0 → overfitting-0.4.1}/PKG-INFO +3 -3
  2. {overfitting-0.4.0 → overfitting-0.4.1}/README.md +2 -2
  3. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/analysis/report.py +29 -25
  4. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/analysis/viewer.py +3 -9
  5. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/broker.py +10 -9
  6. overfitting-0.4.1/overfitting/data.py +89 -0
  7. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/order.py +16 -10
  8. overfitting-0.4.1/overfitting/position.py +113 -0
  9. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/strategy.py +6 -6
  10. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting.egg-info/PKG-INFO +3 -3
  11. {overfitting-0.4.0 → overfitting-0.4.1}/pyproject.toml +1 -1
  12. overfitting-0.4.0/overfitting/data.py +0 -120
  13. overfitting-0.4.0/overfitting/position.py +0 -118
  14. {overfitting-0.4.0 → overfitting-0.4.1}/LICENSE +0 -0
  15. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/__init__.py +0 -0
  16. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/analysis/__init__.py +0 -0
  17. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/analysis/graph.py +0 -0
  18. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/errors.py +0 -0
  19. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/indicators.py +0 -0
  20. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/slippage.py +0 -0
  21. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting/types.py +0 -0
  22. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting.egg-info/SOURCES.txt +0 -0
  23. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting.egg-info/dependency_links.txt +0 -0
  24. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting.egg-info/requires.txt +0 -0
  25. {overfitting-0.4.0 → overfitting-0.4.1}/overfitting.egg-info/top_level.txt +0 -0
  26. {overfitting-0.4.0 → overfitting-0.4.1}/setup.cfg +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: overfitting
3
- Version: 0.4.0
3
+ Version: 0.4.1
4
4
  Summary: A Robust Futures CryptoCurrency Backtesting Library.
5
5
  Author-email: Dohyun Kim <dohyun.k.july@gmail.com>
6
6
  License: MIT
@@ -98,9 +98,9 @@ class MyStrategy(Strategy):
98
98
  if previous_sma_short >= previous_sma_long and sma_short < sma_long and position.qty > 0:
99
99
  self.market_order(self.asset, -position.qty)
100
100
 
101
- backtest_data, benchmark_data = load_data()
101
+ price_df, benchmark_data = load_data()
102
102
  strategy = MyStrategy(
103
- data=backtest_data,
103
+ data={"BTC": price_df},
104
104
  benchmark=benchmark_data, # Default = None Optional
105
105
  initial_capital=100_000, # Default Optional
106
106
  commission_rate=0.0002, # Default Optional
@@ -63,9 +63,9 @@ class MyStrategy(Strategy):
63
63
  if previous_sma_short >= previous_sma_long and sma_short < sma_long and position.qty > 0:
64
64
  self.market_order(self.asset, -position.qty)
65
65
 
66
- backtest_data, benchmark_data = load_data()
66
+ price_df, benchmark_data = load_data()
67
67
  strategy = MyStrategy(
68
- data=backtest_data,
68
+ data={"BTC": price_df},
69
69
  benchmark=benchmark_data, # Default = None Optional
70
70
  initial_capital=100_000, # Default Optional
71
71
  commission_rate=0.0002, # Default Optional
@@ -41,6 +41,7 @@ class Report:
41
41
  self.summary_df, self.drawdown_series, self.drawdown_table = self._compute_summary()
42
42
 
43
43
  def show(self):
44
+ self._print_summary(self.summary_df, self.drawdown_table)
44
45
  self.plot_cumulative() # Plot cumulative returns vs benchmark
45
46
  self.plot_cumulative_log() # Plot cumulative returns on a log scale vs benchmark
46
47
  self.plot_daily_returns() # Plot daily returns over time
@@ -225,24 +226,29 @@ class Report:
225
226
  return gross_returns, return_percents
226
227
 
227
228
  def _compute_summary(self):
228
- peak = self.cum.expanding(min_periods=1).max()
229
- drawdown = (self.cum / peak) - 1
230
- dvt = graph.value_at_risk(self.daily_returns, sigma=2, period=None)
229
+ drawdown = self._compute_drawdown()
230
+ return_stats = self._compute_return_stats(drawdown)
231
+ trade_stats = self._compute_trade_stats()
231
232
 
232
- # monthly stats for skew/kurt
233
- monthly = (1 + self.daily_returns).resample("ME").prod().sub(1)
234
- monthly.index = monthly.index.strftime("%Y-%m")
235
- skew_val = skew(monthly) if len(monthly) > 2 else np.nan
236
- kurt_val = kurtosis(monthly, fisher=False) if len(monthly) > 2 else np.nan
233
+ summary_df = self._format_summary({**return_stats, **trade_stats})
234
+ dd_table = graph.show_worst_drawdown_periods(self.daily_returns)
235
+ return summary_df, drawdown, dd_table
236
+
237
+ def _compute_drawdown(self):
238
+ peak = self.cum.expanding(min_periods=1).max()
239
+ return (self.cum / peak) - 1
237
240
 
241
+ def _compute_return_stats(self, drawdown):
238
242
  years = max(1e-9, (pd.to_datetime(self.end) - pd.to_datetime(self.start)).days / 365)
239
243
  cum_return = float(self.cum.iloc[-1])
240
244
  cagr = (cum_return ** (1 / years)) - 1 if cum_return > 0 else np.nan
241
245
 
242
- sharpe = graph.sharpe_ratio(self.daily_returns, risk_free=0, period="daily")
243
- sortino = graph.sortino_ratio(self.daily_returns, required_return=0, period="daily")
246
+ monthly = (1 + self.daily_returns).resample("ME").prod().sub(1)
247
+ has_monthly = len(monthly) > 2
248
+ skew_val = float(skew(monthly)) if has_monthly else np.nan
249
+ kurt_val = float(kurtosis(monthly, fisher=False)) if has_monthly else np.nan
244
250
 
245
- summary = {
251
+ return {
246
252
  "Number of Years": round(years, 2),
247
253
  "Start Date": self.start,
248
254
  "End Date": self.end,
@@ -250,26 +256,24 @@ class Report:
250
256
  "Final Balance": self.initial_capital * cum_return,
251
257
  "CAGR": cagr,
252
258
  "Cumulative Return": cum_return,
253
- "Sharpe Ratio": sharpe,
254
- "Sortino Ratio": sortino,
259
+ "Sharpe Ratio": graph.sharpe_ratio(self.daily_returns, risk_free=0, period="daily"),
260
+ "Sortino Ratio": graph.sortino_ratio(self.daily_returns, required_return=0, period="daily"),
255
261
  "Max Drawdown": float(drawdown.min()),
256
- "Daily Value At Risk": float(dvt),
257
- "Skew": float(skew_val) if pd.notna(skew_val) else np.nan,
258
- "Kurtosis": float(kurt_val) if pd.notna(kurt_val) else np.nan,
262
+ "Daily Value At Risk": float(graph.value_at_risk(self.daily_returns, sigma=2, period=None)),
263
+ "Skew": skew_val,
264
+ "Kurtosis": kurt_val,
259
265
  }
260
266
 
261
- # trade stats
262
- stats = graph.trade_summary(self.gross_returns, self.return_percents)
263
- summary.update(stats.to_dict())
267
+ def _compute_trade_stats(self):
268
+ return graph.trade_summary(self.gross_returns, self.return_percents).to_dict()
264
269
 
270
+ def _format_summary(self, summary: dict):
265
271
  df = pd.DataFrame.from_dict(summary, orient="index", columns=[""])
266
272
  df[""] = df[""].apply(lambda x: f"{x:,.8f}" if isinstance(x, float) else x)
273
+ return df
267
274
 
275
+ def _print_summary(self, summary_df, dd_table):
268
276
  print("Performance Summary")
269
277
  with pd.option_context("display.colheader_justify", "left", "display.width", None):
270
- print(df.to_string(header=False))
271
-
272
- dd_table = graph.show_worst_drawdown_periods(self.daily_returns)
273
- print(dd_table)
274
-
275
- return df, drawdown, dd_table
278
+ print(summary_df.to_string(header=False))
279
+ print(dd_table)
@@ -4,7 +4,6 @@ import numpy as np
4
4
  import plotly.graph_objects as go
5
5
 
6
6
  from overfitting import Strategy
7
- from overfitting.data import MultiCurrency
8
7
 
9
8
  from PySide6.QtWidgets import (
10
9
  QApplication, QMainWindow, QWidget, QVBoxLayout, QHBoxLayout,
@@ -40,9 +39,7 @@ class StrategyAdapter:
40
39
  self._strategy = strategy
41
40
 
42
41
  def get_symbols(self) -> list[str]:
43
- if isinstance(self._strategy.data, MultiCurrency):
44
- return list(self._strategy.data.symbols)
45
- return ["DEFAULT"]
42
+ return list(self._strategy.data.symbols)
46
43
 
47
44
 
48
45
  class OHLCBuilder:
@@ -78,10 +75,7 @@ class OHLCBuilder:
78
75
  @classmethod
79
76
  def from_strategy(cls, strategy, symbol: str) -> pd.DataFrame:
80
77
  """Extract a tidy OHLC DataFrame for *symbol* from a strategy instance."""
81
- try:
82
- d = strategy.data[symbol] if isinstance(strategy.data, MultiCurrency) else strategy.data
83
- except KeyError:
84
- d = strategy.data
78
+ d = strategy.data[symbol]
85
79
 
86
80
  return pd.DataFrame({
87
81
  "timestamp": pd.to_datetime(d.timestamp),
@@ -114,7 +108,7 @@ class TradeBuilder:
114
108
  if col in df.columns:
115
109
  df[col] = pd.to_datetime(df[col], errors="coerce")
116
110
 
117
- if isinstance(strategy.data, MultiCurrency) and symbol and "symbol" in df.columns:
111
+ if symbol and "symbol" in df.columns:
118
112
  df = df[df["symbol"] == symbol]
119
113
 
120
114
  for c in cls.COLUMNS:
@@ -1,6 +1,7 @@
1
1
  import pandas as pd
2
- from typing import List, Dict, Optional, Union, Tuple
3
- from overfitting.data import Data, MultiCurrency
2
+ from decimal import Decimal
3
+ from typing import List, Dict, Optional, Tuple
4
+ from overfitting.data import Data
4
5
  from overfitting.order import Order
5
6
  from overfitting.position import Position
6
7
  from overfitting.slippage import SlippageModel
@@ -9,8 +10,8 @@ from overfitting.errors import EmptyOrderParameters, InvalidOrderParameters, Liq
9
10
 
10
11
  class Broker:
11
12
  def __init__(self,
12
- data: Union[Data, MultiCurrency],
13
- cash: float,
13
+ data: Data,
14
+ cash: float,
14
15
  commission_rate: float,
15
16
  maint_margin_rate: float,
16
17
  maint_amount:float,
@@ -45,8 +46,8 @@ class Broker:
45
46
  f"positions={list(self.position.keys())}, "
46
47
  f"trades={len(self.trades)})")
47
48
 
48
- def _d(self, symbol: str) -> Data:
49
- return self.data[symbol] if isinstance(self.data, MultiCurrency) else self.data
49
+ def _d(self, symbol: str):
50
+ return self.data[symbol]
50
51
 
51
52
  def _bars(self, symbol: str, i: int) -> Tuple:
52
53
  d = self._d(symbol)
@@ -82,8 +83,8 @@ class Broker:
82
83
  if not symbol or not isinstance(symbol, str):
83
84
  raise InvalidOrderParameters(f"symbol must be a non-empty string. - {symbol}")
84
85
 
85
- if not qty or not isinstance(qty, (int, float)):
86
- raise InvalidOrderParameters(f"qty must be a non-empty float. - {qty}")
86
+ if not qty or not isinstance(qty, (int, float, Decimal)):
87
+ raise InvalidOrderParameters(f"qty must be a non-empty number. - {qty}")
87
88
 
88
89
  if symbol not in self.position:
89
90
  self.position[symbol] = Position(symbol, self.maint_margin_rate, self.maint_amount)
@@ -182,7 +183,7 @@ class Broker:
182
183
 
183
184
  position = self.position[symbol]
184
185
 
185
- notional = abs(order.qty) * order.executed_price
186
+ notional = float(abs(order.qty)) * order.executed_price
186
187
  commission = notional * self.commission_rate
187
188
  pnl = position.process_trade(order, liquidation)
188
189
  executed_time = pd.to_datetime(self.data.index[self._i])
@@ -0,0 +1,89 @@
1
+ import pandas as pd
2
+ import numpy as np
3
+ from types import SimpleNamespace
4
+ from pandas.api.types import is_datetime64_any_dtype, is_integer_dtype, is_float_dtype
5
+ from overfitting.errors import InitializationError
6
+
7
+ REQUIRED_OHLC = ("open", "high", "low", "close")
8
+
9
+ class Data(dict):
10
+ """
11
+ data = Data({"BTCUSDT": df_btc, "ETHUSDT": df_eth})
12
+ data["BTCUSDT"].close[i] # close price of BTC at step i
13
+ data.index[i] # the timestamp at step i
14
+ data.symbols # ("BTCUSDT", "ETHUSDT")
15
+ len(data) # how many bars there are
16
+ """
17
+ def __init__(self, frames: dict[str, pd.DataFrame]):
18
+ if not isinstance(frames, dict) or not frames:
19
+ raise InitializationError("Pass a non-empty dict mapping symbol -> DataFrame.")
20
+
21
+ bars = {symbol: _bars_from_frame(df, symbol) for symbol, df in frames.items()}
22
+ _check_aligned_timelines(bars)
23
+
24
+ super().__init__(bars)
25
+ self.symbols = tuple(bars)
26
+ self.index = next(iter(bars.values())).timestamp
27
+
28
+ def __len__(self) -> int:
29
+ return len(self.index)
30
+
31
+
32
+ def _bars_from_frame(df: pd.DataFrame, symbol: str) -> SimpleNamespace:
33
+ """Turn one symbol's DataFrame into a namespace of numpy arrays."""
34
+ if not isinstance(df, pd.DataFrame) or df.empty:
35
+ raise InitializationError(f"'{symbol}': expected a non-empty DataFrame.")
36
+
37
+ missing = [c for c in REQUIRED_OHLC if c not in df.columns]
38
+ if missing:
39
+ raise InitializationError(
40
+ f"'{symbol}': missing required columns {missing}. Got: {list(df.columns)}"
41
+ )
42
+
43
+ timestamps = _pull_timestamps(df, symbol)
44
+
45
+ arrays = {col: df[col].to_numpy() for col in df.columns}
46
+ arrays["timestamp"] = timestamps
47
+ return SimpleNamespace(**arrays)
48
+
49
+
50
+ def _pull_timestamps(df: pd.DataFrame, symbol: str) -> np.ndarray:
51
+ """Find the timestamp series — either a column or the DatetimeIndex — and normalise it."""
52
+ if "timestamp" in df.columns:
53
+ ts = df["timestamp"]
54
+ elif isinstance(df.index, pd.DatetimeIndex) or df.index.name == "timestamp":
55
+ ts = pd.Series(df.index, name="timestamp")
56
+ else:
57
+ raise InitializationError(
58
+ f"'{symbol}': add a 'timestamp' column or use a DatetimeIndex."
59
+ )
60
+ return _to_datetime_ns(ts).to_numpy()
61
+
62
+
63
+ def _to_datetime_ns(ts: pd.Series) -> pd.Series:
64
+ """Coerce whatever the user gave us into tz-naive datetime64[ns]."""
65
+ if is_datetime64_any_dtype(ts):
66
+ dt = pd.to_datetime(ts)
67
+ try:
68
+ return dt.tz_localize(None)
69
+ except Exception:
70
+ return dt
71
+
72
+ if is_integer_dtype(ts) or is_float_dtype(ts):
73
+ # Big numbers look like milliseconds; smaller ones look like seconds.
74
+ unit = "ms" if pd.Series(ts).max() > 1e12 else "s"
75
+ return pd.to_datetime(ts, unit=unit)
76
+
77
+ return pd.to_datetime(ts, errors="raise")
78
+
79
+
80
+ def _check_aligned_timelines(bars: dict[str, SimpleNamespace]) -> None:
81
+ """All symbols must share the same timeline — for now."""
82
+ # TODO: auto-fill / reindex so symbols with different timelines can co-exist.
83
+ iterator = iter(bars.items())
84
+ _, first = next(iterator)
85
+ for symbol, b in iterator:
86
+ if len(b.timestamp) != len(first.timestamp) or not np.array_equal(b.timestamp, first.timestamp):
87
+ raise InitializationError(
88
+ f"'{symbol}' timeline doesn't match the other symbols."
89
+ )
@@ -1,23 +1,25 @@
1
1
  import uuid
2
+ from decimal import Decimal
2
3
  import pandas as pd
3
4
  from overfitting.types import OrderType
4
5
 
5
6
  class Order:
6
- def __init__(self,
7
- time: pd.Timestamp,
8
- symbol: str,
9
- qty: float,
10
- price:float,
11
- type: OrderType,
7
+ def __init__(self,
8
+ time: pd.Timestamp,
9
+ symbol: str,
10
+ qty: float,
11
+ price:float,
12
+ type: OrderType,
12
13
  stop_price: float = None,
13
14
  label: str= None):
14
-
15
+
15
16
  self.id = uuid.uuid4().hex[:16]
16
17
  self.created_at = time
17
18
  self.executed_at = None
18
19
  self.symbol = symbol
19
- self.side = "LONG" if qty > 0 else "SHORT"
20
- self.qty = qty
20
+ # Decimal qty so Position bookkeeping is exact
21
+ self.qty = Decimal(str(qty))
22
+ self.side = "LONG" if self.qty > 0 else "SHORT"
21
23
  self.price = price
22
24
  self.type = type
23
25
  self.stop_price = stop_price
@@ -50,7 +52,11 @@ class Order:
50
52
 
51
53
  def to_dict(self):
52
54
  return {
53
- k: (v.name if isinstance(v, OrderType) else v)
55
+ k: (
56
+ v.name if isinstance(v, OrderType)
57
+ else float(v) if isinstance(v, Decimal)
58
+ else v
59
+ )
54
60
  for k, v in self.__dict__.items()
55
61
  }
56
62
 
@@ -0,0 +1,113 @@
1
+ from decimal import Decimal
2
+ from overfitting.order import Order
3
+
4
+ ZERO = Decimal("0")
5
+
6
+ class Position:
7
+ def __init__(self, symbol: str, maint_margin_rate: float = 0.005, maint_amount: float = 0):
8
+ self.symbol = symbol
9
+ self.maint_margin_rate = maint_margin_rate
10
+ self.maint_amount = maint_amount
11
+ self.leverage = 1
12
+
13
+ # Open-position state. All zero when flat.
14
+ self.qty = ZERO
15
+ self.price = 0.0
16
+ self.margin = 0.0
17
+ self.liquid_price = 0.0
18
+
19
+ def __repr__(self):
20
+ return (f"Position(symbol={self.symbol}, qty={self.qty}, price={self.price}, "
21
+ f"liquid_price={self.liquid_price}, margin={self.margin}, "
22
+ f"leverage={self.leverage})")
23
+
24
+ def _recalc_liquidation(self) -> None:
25
+ """
26
+ Liquidation (isolated margin):
27
+ notional = abs(qty) * entry_price
28
+ initial_margin = notional / leverage
29
+ maint_margin = max(0, notional * maint_margin_rate - maint_amount)
30
+
31
+ long liquidation = entry - (initial_margin - maint_margin) / abs(qty)
32
+ short liquidation = entry + (initial_margin - maint_margin) / abs(qty)
33
+ """
34
+ size = abs(self.qty)
35
+ if size == 0:
36
+ # Flat — no margin and no liquidation price to track.
37
+ self.margin = 0.0
38
+ self.liquid_price = 0.0
39
+ return
40
+
41
+ # qty is Decimal; cast to float for the price-side math.
42
+ size_f = float(size)
43
+
44
+ # Isolated-margin model.
45
+ # initial_margin: what we lock up to open the position at this leverage.
46
+ # maint_margin: cushion the exchange requires before forcing liquidation.
47
+ notional = self.price * size_f
48
+ initial_margin = notional / self.leverage
49
+ maint_margin = max(0.0, notional * self.maint_margin_rate - self.maint_amount)
50
+
51
+ self.margin = initial_margin + maint_margin
52
+
53
+ # How far the price can move against us before we get liquidated.
54
+ # Long: liquidates when price drops by `offset`.
55
+ # Short: liquidates when price rises by `offset`.
56
+ offset = (initial_margin - maint_margin) / size_f
57
+ self.liquid_price = self.price - offset if self.qty > 0 else self.price + offset
58
+
59
+ def _add_to_position(self, order: Order) -> None:
60
+ new_qty = self.qty + order.qty
61
+ # Weighted-average entry price
62
+ self.price = (
63
+ self.price * float(self.qty) + order.executed_price * float(order.qty)
64
+ ) / float(new_qty)
65
+ self.qty = new_qty
66
+ self._recalc_liquidation()
67
+
68
+ def _reset(self) -> None:
69
+ self.qty = ZERO
70
+ self.price = 0.0
71
+ self.margin = 0.0
72
+ self.liquid_price = 0.0
73
+
74
+ def set_leverage(self, leverage: int) -> None:
75
+ if leverage <= 0 or leverage > 100:
76
+ raise Exception("Leverage must be between 1 and 100.")
77
+ self.leverage = leverage
78
+ self._recalc_liquidation()
79
+
80
+ def process_trade(self, order: Order, liquidation: bool = False) -> float:
81
+ """Apply an executed order. Returns realized PnL (commission excluded)."""
82
+ if order.symbol != self.symbol:
83
+ raise ValueError(f"Order symbol {order.symbol!r} doesn't match position {self.symbol!r}.")
84
+ if order.qty == 0:
85
+ raise ValueError("Order quantity cannot be zero.")
86
+
87
+ if liquidation: # Force liquidation
88
+ loss = -self.margin
89
+ self._reset()
90
+ return loss
91
+
92
+ # Flat or same-side add: update entry price and increase position size.
93
+ if self.qty == 0 or self.qty * order.qty > 0:
94
+ self._add_to_position(order)
95
+ return 0.0
96
+ else:
97
+ closed_qty = min(abs(self.qty), abs(order.qty))
98
+ direction = 1 if self.qty > 0 else -1
99
+ # Long earns (exit - entry) and short earns (entry - exit). Same formula with sign flip.
100
+ pnl = direction * (order.executed_price - self.price) * float(closed_qty)
101
+
102
+ self.qty += order.qty
103
+
104
+ if self.qty == 0:
105
+ self._reset()
106
+ else:
107
+ # If we flipped sides, the new entry price is the executed price.
108
+ flipped = self.qty * order.qty > 0
109
+ if flipped:
110
+ self.price = order.executed_price
111
+ self._recalc_liquidation()
112
+
113
+ return pnl
@@ -2,8 +2,8 @@ import os
2
2
  import pandas as pd
3
3
  import numpy as np
4
4
  from abc import abstractmethod
5
- from typing import List, Optional, Union, Dict
6
- from overfitting.data import Data, MultiCurrency
5
+ from typing import List, Optional, Dict
6
+ from overfitting.data import Data
7
7
  from overfitting.broker import Broker
8
8
  from overfitting.order import Order
9
9
  from overfitting.position import Position
@@ -11,8 +11,8 @@ from overfitting.analysis.report import Report
11
11
  from overfitting.slippage import SlippageModel
12
12
 
13
13
  class Strategy:
14
- def __init__(self,
15
- data: Union[pd.DataFrame, Dict[str, pd.DataFrame]],
14
+ def __init__(self,
15
+ data: Dict[str, pd.DataFrame],
16
16
  *,
17
17
  benchmark: Optional[pd.DataFrame] = None,
18
18
  initial_capital: float =100000,
@@ -20,9 +20,9 @@ class Strategy:
20
20
  maint_margin_rate: float =0.005,
21
21
  maint_amount: float=0,
22
22
  slippage_model: Optional[SlippageModel] = None):
23
-
23
+
24
24
  self.benchmark = benchmark
25
- self.data = MultiCurrency(data) if isinstance(data, dict) else Data(data)
25
+ self.data = Data(data)
26
26
  self.broker = Broker(
27
27
  data=self.data,
28
28
  cash=initial_capital,
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: overfitting
3
- Version: 0.4.0
3
+ Version: 0.4.1
4
4
  Summary: A Robust Futures CryptoCurrency Backtesting Library.
5
5
  Author-email: Dohyun Kim <dohyun.k.july@gmail.com>
6
6
  License: MIT
@@ -98,9 +98,9 @@ class MyStrategy(Strategy):
98
98
  if previous_sma_short >= previous_sma_long and sma_short < sma_long and position.qty > 0:
99
99
  self.market_order(self.asset, -position.qty)
100
100
 
101
- backtest_data, benchmark_data = load_data()
101
+ price_df, benchmark_data = load_data()
102
102
  strategy = MyStrategy(
103
- data=backtest_data,
103
+ data={"BTC": price_df},
104
104
  benchmark=benchmark_data, # Default = None Optional
105
105
  initial_capital=100_000, # Default Optional
106
106
  commission_rate=0.0002, # Default Optional
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
4
4
 
5
5
  [project]
6
6
  name = "overfitting"
7
- version = "0.4.0"
7
+ version = "0.4.1"
8
8
  description = "A Robust Futures CryptoCurrency Backtesting Library."
9
9
  readme = "README.md"
10
10
  requires-python = ">=3.8"
@@ -1,120 +0,0 @@
1
- import pandas as pd
2
- import numpy as np
3
- from pandas.api.types import is_datetime64_any_dtype, is_integer_dtype, is_float_dtype
4
- from overfitting.errors import InitializationError
5
-
6
- REQUIRED_OHLC = ("open", "high", "low", "close")
7
-
8
- class Data(dict):
9
-
10
- open: np.ndarray # [Required] Open prices
11
- high: np.ndarray # [Required] High prices
12
- low: np.ndarray # [Required] Low prices
13
- close: np.ndarray # [Required] Close prices
14
- timestamp: np.ndarray # Candle timestamps
15
- index: np.ndarray # Fast datetime64[ns] index
16
- columns: tuple[str, ...] # Available columns
17
- n: int # Number of rows
18
-
19
- """
20
- Usage: data.open[i], data.high[i], data.timestamp[i]
21
- """
22
-
23
- def __init__(self, df: pd.DataFrame):
24
- if not isinstance(df, pd.DataFrame) or df.empty:
25
- raise InitializationError("Data must be a non-empty pandas DataFrame.")
26
-
27
- # Validate OHLC
28
- missing = [c for c in REQUIRED_OHLC if c not in df.columns]
29
- if missing:
30
- raise InitializationError(f"Missing required columns: {missing}. Available: {list(df.columns)}")
31
-
32
- # Resolve timestamp: column or index
33
- if "timestamp" in df.columns:
34
- ts = df["timestamp"]
35
- elif isinstance(df.index, pd.DatetimeIndex) or df.index.name == "timestamp":
36
- # normalize to a Series so downstream is uniform
37
- ts = pd.Series(df.index, index=df.index, name="timestamp")
38
- else:
39
- raise InitializationError("Provide a 'timestamp' column or use a DatetimeIndex.")
40
-
41
- # Coerce timestamp to datetime64[ns]
42
- ts = self._to_datetime_ns(ts)
43
-
44
- payload = {c: df[c].to_numpy() for c in df.columns}
45
- payload["timestamp"] = ts.to_numpy()
46
-
47
- super().__init__(payload)
48
- object.__setattr__(self, "columns", tuple(sorted(payload.keys(), key=lambda x: (x!='timestamp', x))))
49
- object.__setattr__(self, "index", ts.to_numpy())
50
- object.__setattr__(self, "n", len(df))
51
-
52
- @staticmethod
53
- def _to_datetime_ns(s: pd.Series) -> pd.Series:
54
- if is_datetime64_any_dtype(s):
55
- dt = pd.to_datetime(s)
56
- # drop tz if present
57
- try:
58
- dt = dt.tz_localize(None)
59
- except Exception:
60
- pass
61
- return dt
62
-
63
- if is_integer_dtype(s) or is_float_dtype(s):
64
- # Heuristic: ms since epoch vs seconds
65
- mx = pd.Series(s).max()
66
- unit = "ms" if mx > 1e12 else "s"
67
- return pd.to_datetime(s, unit=unit)
68
-
69
- # Strings or mixed -> let pandas parse
70
- return pd.to_datetime(s, errors="raise")
71
-
72
- def __getattr__(self, key):
73
- # attribute-style access for columns
74
- try:
75
- return self[key]
76
- except KeyError:
77
- raise AttributeError( f"'Data' has no field '{key}'. Available: {', '.join(self.columns)}")
78
-
79
- def __len__(self):
80
- return self.n
81
-
82
- def __setattr__(self, key, value):
83
- # keep the container read-only (avoids accidental mistakes)
84
- if key in {"index", "columns", "n"}:
85
- object.__setattr__(self, key, value)
86
- else:
87
- raise AttributeError("Data is read-only; modify the source DataFrame before wrapping.")
88
-
89
- class MultiCurrency(dict):
90
-
91
- symbols: tuple[str, ...] # List of symbols in this container
92
- index: np.ndarray # datetime64[ns] index
93
- n: int # Number of rows per symbol
94
-
95
- def __init__(self, frames: dict[str, pd.DataFrame]):
96
- if not isinstance(frames, dict) or not frames:
97
- raise InitializationError("Data must be non-empty dict Type - dict[str, pd.DataFrame]")
98
-
99
- payload = {}
100
- first_ts = None
101
- for symbol, df in frames.items():
102
- d = Data(df)
103
- if first_ts is None:
104
- first_ts = d.index
105
- else:
106
- # Requires identical timestamps for NOW
107
- # TODO IMNPLEMENT AUTO FILL for uniform timestamps
108
- if len(d.index) != len(first_ts) or not np.array_equal(d.index, first_ts):
109
- raise InitializationError(
110
- f"Len Timestamps for {symbol} are not equal with the other symbols.")
111
-
112
- payload[symbol] = d
113
-
114
- super().__init__(payload)
115
- object.__setattr__(self, "symbols", tuple(payload.keys()))
116
- object.__setattr__(self, "index", first_ts)
117
- object.__setattr__(self, "n", len(first_ts))
118
-
119
- def __len__(self) -> int:
120
- return self.n
@@ -1,118 +0,0 @@
1
- from math import copysign
2
- from overfitting.order import Order
3
-
4
- class Position:
5
- def __init__(self,
6
- symbol:str =None,
7
- maint_margin_rate:float=0.005,
8
- maint_amount:float=0):
9
-
10
- self.symbol = symbol
11
- self.qty = 0.0
12
- self.price = 0.0
13
- self.liquid_price = 0.0
14
- self.margin = 0.0
15
- self.leverage = 1 # Default Leverage
16
- self.maint_margin_rate = maint_margin_rate
17
- self.maint_amount = maint_amount
18
-
19
- def __repr__(self):
20
- return (f"Position("
21
- f"symbol={self.symbol}, "
22
- f"qty={self.qty}, "
23
- f"price={self.price}, "
24
- f"liquid_price={self.liquid_price}, "
25
- f"margin={self.margin}, "
26
- f"leverage={self.leverage}, "
27
- f"maint_margin_rate={self.maint_margin_rate}, "
28
- f"maint_amount={self.maint_amount})")
29
-
30
- def _update_liquid_price(self):
31
- """
32
- NOTE: liquidation price is calculated based on ISOLATED mode.
33
-
34
- Liquidation Price Calculation:
35
- * Initial Margin = price * size / leverage
36
- * Maint Margin = price * size * margin rate - margin amount
37
- [LONG] LP = Entry Price - (Initial Margin - Maintenance Margin)
38
- [SHORT] LP = Entry Price + (Initial Margin - Maintenance Margin)
39
- """
40
- q = abs(self.qty)
41
- if q == 0:
42
- self.liquid_price = 0.0
43
- self.margin = 0.0
44
- return
45
-
46
- notional = self.price * q
47
- im = notional / self.leverage
48
- mm = max(0.0, notional * self.maint_margin_rate - self.maint_amount)
49
-
50
- self.margin = im + mm
51
- delta_p = (im - mm) / q
52
-
53
- if self.qty > 0: # long
54
- self.liquid_price = self.price - delta_p
55
- else: # short
56
- self.liquid_price = self.price + delta_p
57
-
58
- def _liquidate(self):
59
- """Returns PNL which is position margin * -1"""
60
- pnl = -self.margin
61
- self._reset()
62
- return pnl
63
-
64
- def _reset(self):
65
- self.qty = 0.0
66
- self.price = 0.0
67
- self.liquid_price = 0.0
68
- self.margin = 0.0
69
-
70
- def set_leverage(self, leverage):
71
- if leverage <= 0 or leverage > 100:
72
- raise Exception("set_leverage() Invalid Leverage. Please Choose Between 0 and 100")
73
-
74
- self.leverage = leverage
75
- self._update_liquid_price()
76
-
77
- def _calculate_realized_pnl(self, entry_price: float, exit_price: float, qty: float, side: float):
78
- """
79
- side > 0 : position is long
80
- side < 0 : position is short
81
- """
82
- if side > 0:
83
- return (exit_price - entry_price) * qty
84
- return (entry_price - exit_price) * qty
85
-
86
- def process_trade(self, order: Order, liquidation: bool = False) -> float:
87
- if self.symbol != order.symbol:
88
- raise ValueError("Cannot process trade with a different symbol.")
89
- if order.qty == 0:
90
- raise ValueError("Transaction quantity cannot be zero.")
91
-
92
- if liquidation:
93
- return self._liquidate()
94
-
95
- same_side = self.qty == 0 or (self.qty * order.qty > 0)
96
-
97
- if same_side:
98
- total = self.qty + order.qty
99
- self.price = (self.price * self.qty + order.executed_price * order.qty) / total
100
- self.qty = total
101
- self._update_liquid_price()
102
- return 0.0
103
-
104
- # Opposite side — reduce, close, or flip
105
- close_qty = min(abs(self.qty), abs(order.qty))
106
- side = 1 if self.qty > 0 else -1
107
- pnl = self._calculate_realized_pnl(self.price, order.executed_price, close_qty, side)
108
-
109
- self.qty += order.qty
110
-
111
- if self.qty == 0:
112
- self._reset()
113
- else: # flip or reduce
114
- if self.qty * order.qty > 0: # flipped
115
- self.price = order.executed_price
116
- self._update_liquid_price()
117
-
118
- return pnl
File without changes
File without changes