siglab-py 0.1.29__py3-none-any.whl → 0.6.12__py3-none-any.whl

siglab_py/ordergateway/test_ordergateway.py

@@ -12,7 +12,7 @@ from constants import JSON_SERIALIZABLE_TYPES
 
 '''
 set PYTHONPATH=%PYTHONPATH%;D:\dev\siglab\siglab_py
-python test_ordergateway.py --gateway_id bybit_03
+python test_ordergateway.py --gateway_id hyperliquid_01
 '''
 
 param : Dict[str, str] = {
@@ -91,27 +91,28 @@ if __name__ == '__main__':
 
     positions_3 : List[DivisiblePosition] = [
         DivisiblePosition(
-            ticker = 'BTC/USDT:USDT',
+            ticker = 'BTC/USDC:USDC',
             side = 'sell',
-            amount = 0.01,
+            amount = 0.00100,
             leg_room_bps = 5,
-            reduce_only=False,
+            reduce_only=True,
             order_type = 'limit',
             slices=1,
-            wait_fill_threshold_ms=60000
+            wait_fill_threshold_ms=60000,
+            fees_ccy='USDC'
         )
     ]
 
     
     executed_positions : Union[Dict[JSON_SERIALIZABLE_TYPES, JSON_SERIALIZABLE_TYPES], None] = execute_positions(
         redis_client=redis_client,
-        positions=positions_2,
+        positions=positions_3,
         ordergateway_pending_orders_topic=ordergateway_pending_orders_topic,
         ordergateway_executions_topic=ordergateway_executions_topic
         )
     if executed_positions:
         for position in executed_positions:
-            print(f"{position['ticker']} {position['side']} amount: {position['amount']} leg_room_bps: {position['leg_room_bps']} slices: {position['slices']}, filled_amount: {position['filled_amount']}, average_cost: {position['average_cost']}, # executions: {len(position['executions'])}") # type: ignore
+            print(f"{position['ticker']} {position['side']} amount: {position['amount']} leg_room_bps: {position['leg_room_bps']} slices: {position['slices']}, filled_amount: {position['filled_amount']}, average_cost: {position['average_cost']}, # executions: {len(position['executions'])}, done: {position['done']}, execution_err: {position['execution_err']}") # type: ignore
         
         
 

siglab_py/tests/integration/market_data_util_tests.py

@@ -100,7 +100,7 @@ class MarketDataUtilTests(unittest.TestCase):
                 'defaultType': 'swap'            }
         }
 
-        exchange : Exchange = okx(param)
+        exchange : Exchange = okx(param) # type: ignore
         normalized_symbols = [ 'BTC/USDT:USDT' ]
         pd_candles: Union[pd.DataFrame, None] = fetch_candles(
             start_ts=start_date.timestamp(),
@@ -118,6 +118,40 @@ class MarketDataUtilTests(unittest.TestCase):
             assert set(pd_candles.columns) >= expected_columns, "Missing expected columns."
             assert pd_candles['timestamp_ms'].notna().all(), "timestamp_ms column contains NaN values."
             assert pd_candles['timestamp_ms'].is_monotonic_increasing, "Timestamps are not in ascending order."
+
+    def test_aggregate_candles(self):
+        end_date : datetime = datetime.today()
+        start_date : datetime = end_date + timedelta(hours=-8)
+
+        param = {
+            'apiKey' : None,
+            'secret' : None,
+            'password' : None,
+            'subaccount' : None,
+            'rateLimit' : 100,    # In ms
+            'options' : {
+                'defaultType': 'swap'            }
+        }
+
+        exchange : Exchange = okx(param) # type: ignore
+        normalized_symbols = [ 'BTC/USDT:USDT' ]
+        pd_candles: Union[pd.DataFrame, None] = fetch_candles(
+            start_ts=start_date.timestamp(),
+            end_ts=end_date.timestamp(),
+            exchange=exchange,
+            normalized_symbols=normalized_symbols,
+            candle_size='15m' # <---- aggregate 1m into 15m candles
+        )[normalized_symbols[0]]
+
+        assert pd_candles is not None
+        pd_candles['timestamp_ms_gap'] = pd_candles['timestamp_ms'].diff()
+        timestamp_ms_gap_median = pd_candles['timestamp_ms_gap'].median()
+        NUM_MS_IN_1HR = 60*60*1000
+        expected_15m_gap_ms = NUM_MS_IN_1HR/4
+        assert(timestamp_ms_gap_median==expected_15m_gap_ms)
+        total_num_rows = pd_candles.shape[0]
+        num_rows_with_15min_gaps = pd_candles[pd_candles.timestamp_ms_gap!=timestamp_ms_gap_median].shape[0]
+        assert(num_rows_with_15min_gaps/total_num_rows <= 0.4) # Why not 100% match? minute bars may have gaps (Also depends on what ticker)
         
     def test_fetch_candles_futubull(self):
         # You need Futu OpenD running and you need entitlements

siglab_py/tests/unit/analytic_util_tests.py

@@ -2,7 +2,7 @@ import unittest
 from typing import List
 from pathlib import Path
 
-from util.analytic_util import compute_candles_stats
+from util.analytic_util import compute_candles_stats, lookup_fib_target
 
 import pandas as pd
 
@@ -38,31 +38,41 @@ class AnalyticUtilTests(unittest.TestCase):
             pd_candles=pd_candles,
             boillenger_std_multiples=2,
             sliding_window_how_many_candles=20,
-            pypy_compat=True
+            pypy_compat=True # Slopes calculation? Set pypy_compat to False
         )
 
         expected_columns : List[str] = [
             'exchange', 'symbol', 'timestamp_ms', 
             'open', 'high', 'low', 'close', 'volume', 
             'datetime', 'datetime_utc', 'year', 'month', 'day', 'hour', 'minute', 'dayofweek', 
-            'pct_chg_on_close', 'candle_height', 
+            'pct_chg_on_close', 'candle_height', 'candle_body_height',
             'week_of_month', 'apac_trading_hr', 'emea_trading_hr', 'amer_trading_hr', 
-            'is_green', 'pct_change_close', 
+            'is_green', 'candle_class', 'pct_change_close', 
             'sma_short_periods', 'sma_long_periods', 'ema_short_periods', 'ema_long_periods', 'ema_close', 
-            'std', 'std_percent', 'candle_height_percent', 'candle_height_percent_rounded', 
+            'std', 'std_percent', 
+            'vwap_short_periods', 'vwap_long_periods',
+            'candle_height_percent', 'candle_height_percent_rounded', 'candle_body_height_percent', 'candle_body_height_percent_rounded',
+            'log_return', 'interval_hist_vol', 'annualized_hist_vol',
             'chop_against_ema', 
             'ema_volume_short_periods', 'ema_volume_long_periods', 
+            'ema_cross', 'ema_cross_last', 'ema_bullish_cross_last_id', 'ema_bearish_cross_last_id',
             'max_short_periods', 'max_long_periods', 'idmax_short_periods', 'idmax_long_periods', 'min_short_periods', 'min_long_periods', 'idmin_short_periods', 'idmin_long_periods', 
-            'h_l', 'h_pc', 'l_pc', 'tr', 'atr', 
+            'max_candle_body_height_percent_long_periods', 'idmax_candle_body_height_percent_long_periods',
+            'min_candle_body_height_percent_long_periods', 'idmin_candle_body_height_percent_long_periods',
+            'price_swing_short_periods', 'price_swing_long_periods',
+            'trend_from_highs_long_periods', 'trend_from_lows_long_periods', 'trend_from_highs_short_periods', 'trend_from_lows_short_periods',
+            'h_l', 'h_pc', 'l_pc', 'tr', 'atr', 'atr_avg_short_periods', 'atr_avg_long_periods',
             'hurst_exp', 
             'boillenger_upper', 'boillenger_lower', 'boillenger_channel_height', 'boillenger_upper_agg', 'boillenger_lower_agg', 'boillenger_channel_height_agg', 
             'aggressive_up', 'aggressive_up_index', 'aggressive_up_candle_height', 'aggressive_up_candle_high', 'aggressive_up_candle_low', 'aggressive_down', 'aggressive_down_index', 'aggressive_down_candle_height', 'aggressive_down_candle_high', 'aggressive_down_candle_low', 
             'fvg_low', 'fvg_high', 'fvg_gap', 'fvg_mitigated', 
-            'close_delta', 'close_delta_percent', 'up', 'down', 'rsi', 'ema_rsi', 'typical_price', 
-            'money_flow', 'money_flow_positive', 'money_flow_negative', 'positive_flow_sum', 'negative_flow_sum', 'money_flow_ratio', 'mfi', 
-            'macd', 'signal', 'macd_minus_signal', 
-            'fib_618_short_periods', 'fib_618_long_periods', 
-            'gap_close_vs_ema', 
+            'close_delta', 'close_delta_percent', 'up', 'down', 
+            'rsi', 'rsi_bucket', 'ema_rsi', 'rsi_max', 'rsi_idmax', 'rsi_min', 'rsi_idmin', 'rsi_trend', 'rsi_trend_from_highs', 'rsi_trend_from_lows', 'rsi_divergence',
+            'typical_price', 
+            'money_flow', 'money_flow_positive', 'money_flow_negative', 'positive_flow_sum', 'negative_flow_sum', 'money_flow_ratio', 'mfi', 'mfi_bucket',
+            'macd', 'signal', 'macd_minus_signal', 'macd_cross', 'macd_bullish_cross_last_id', 'macd_bearish_cross_last_id', 'macd_cross_last',
+            'fib_0.618_short_periods', 'fib_0.618_long_periods', 
+            'gap_close_vs_ema', 'gap_close_vs_ema_percent',
             'close_above_or_below_ema', 
             'close_vs_ema_inflection'
         ]
@@ -70,4 +80,29 @@ class AnalyticUtilTests(unittest.TestCase):
         missing_columns = [ expected for expected in expected_columns if expected not in pd_candles.columns.to_list() ]
         unexpected_columns = [ actual for actual in pd_candles.columns.to_list() if actual not in expected_columns ]
 
-        assert(pd_candles.columns.to_list()==expected_columns)
+        assert(pd_candles.columns.to_list()==expected_columns)
+
+    def test_lookup_fib_target(self):
+        data_dir = Path(__file__).parent.parent.parent.parent / "data"
+        csv_path = data_dir / "sample_btc_candles.csv"
+        pd_candles : pd.DataFrame = pd.read_csv(csv_path)
+        target_fib_level : float = 0.618
+        compute_candles_stats(
+            pd_candles=pd_candles,
+            boillenger_std_multiples=2,
+            sliding_window_how_many_candles=20,
+            target_fib_level=target_fib_level,
+            pypy_compat=True # Slopes calculation? Set pypy_compat to False
+        )
+
+        last_row = pd_candles.iloc[-1]
+        result = lookup_fib_target(
+            row=last_row, 
+            pd_candles=pd_candles,
+            target_fib_level=target_fib_level
+            )
+        if result:
+            assert(result['short_periods']['min']<result['short_periods']['fib_target']<result['short_periods']['max'])
+            assert(result['long_periods']['min']<result['long_periods']['fib_target']<result['long_periods']['max'])
+        
+

siglab_py/tests/unit/simple_math_tests.py

@@ -0,0 +1,235 @@
+import unittest
+from typing import List, Dict, Union
+
+from util.simple_math import generate_rand_nums, round_to_level, bucket_series, bucketize_val
+
+class SimpleMathTests(unittest.TestCase):
+
+    def test_generate_rand_nums(self):
+        range_min : float = 0
+        range_max : float = 1
+        size : int = 100
+        percentage_in_range : float = 91
+        abs_min : float = -0.5
+        abs_max : float = 1.1
+
+        rand_nums : List[float] = generate_rand_nums(
+                                        range_min = range_min,
+                                        range_max = range_max,
+                                        size = size, 
+                                        percent_in_range = percentage_in_range,
+                                        abs_min = abs_min,
+                                        abs_max = abs_max
+                                    )
+        
+        assert(len(rand_nums)==size)
+        assert(len([x for x in rand_nums if x>=range_min and x<=range_max]) == (percentage_in_range/100) * size)
+        assert(len([x for x in rand_nums if x<abs_min or x>abs_max]) == 0)
+
+
+        range_min = -1
+        range_max  = 1
+        percentage_in_range = 91
+        abs_min = -1.5
+        abs_max = 1.5
+
+        rand_nums : List[float] = generate_rand_nums(
+                                        range_min = range_min,
+                                        range_max = range_max,
+                                        size = size, 
+                                        percent_in_range = percentage_in_range,
+                                        abs_min = abs_min,
+                                        abs_max = abs_max
+                                    )
+        
+        assert(len(rand_nums)==size)
+        assert(len([x for x in rand_nums if x>=range_min and x<=range_max]) == (percentage_in_range/100) * size)
+        assert(len([x for x in rand_nums if x<abs_min or x>abs_max]) == 0)
+
+
+        range_min = 0
+        range_max  = 100
+        percentage_in_range = 91
+        abs_min = -150
+        abs_max = 150
+
+        rand_nums : List[float] = generate_rand_nums(
+                                        range_min = range_min,
+                                        range_max = range_max,
+                                        size = size, 
+                                        percent_in_range = percentage_in_range,
+                                        abs_min = abs_min,
+                                        abs_max = abs_max
+                                    )
+        
+        assert(len(rand_nums)==size)
+        assert(len([x for x in rand_nums if x>=range_min and x<=range_max]) == (percentage_in_range/100) * size)
+        assert(len([x for x in rand_nums if x<abs_min or x>abs_max]) == 0)
+
+
+        range_min = -100
+        range_max  = 100
+        percentage_in_range = 91
+        abs_min = -150
+        abs_max = 150
+
+        rand_nums : List[float] = generate_rand_nums(
+                                        range_min = range_min,
+                                        range_max = range_max,
+                                        size = size, 
+                                        percent_in_range = percentage_in_range,
+                                        abs_min = abs_min,
+                                        abs_max = abs_max
+                                    )
+        
+        assert(len(rand_nums)==size)
+        assert(len([x for x in rand_nums if x>=range_min and x<=range_max]) == (percentage_in_range/100) * size)
+        assert(len([x for x in rand_nums if x<abs_min or x>abs_max]) == 0)
+
+    def test_round_to_level(self):
+        prices = [ 
+            { 'price' : 15080, 'rounded' : 15000}, 
+            { 'price' : 15180, 'rounded' : 15200}, 
+            { 'price' : 25080, 'rounded' : 25200}, 
+            { 'price' : 25180, 'rounded' : 25200}, 
+            { 'price' : 25380, 'rounded' : 25500}, 
+            { 'price' : 95332, 'rounded' : 95000}, 
+            { 'price' : 95878, 'rounded' : 96000}, 
+            { 'price' : 103499, 'rounded' : 103000}, 
+            { 'price' : 103500, 'rounded' : 104000}, 
+            { 'price' : 150800, 'rounded' : 150000}, 
+            { 'price' : 151800, 'rounded' : 152000}
+        ]
+        for entry in prices:
+            price = entry['price']
+            expected = entry['rounded']
+            rounded_price = round_to_level(price, level_granularity=0.01)
+            print(f"{price} rounded to: {rounded_price}")
+            assert(rounded_price==expected)
+
+    def test_bucket_series(self):
+
+        level_granularity : float = 0.1
+
+        range_min : float = 0
+        range_max : float = 1
+        size : int = 100
+        percentage_in_range : float = 91
+        abs_min : float = -0.5
+        abs_max : float = 1.1
+
+        rand_nums : List[float] = generate_rand_nums(
+                                        range_min = range_min,
+                                        range_max = range_max,
+                                        size = size, 
+                                        percent_in_range = percentage_in_range,
+                                        abs_min = abs_min,
+                                        abs_max = abs_max
+                                    )
+
+        buckets : Dict[
+            str, 
+            Dict[str,Union[float, List[float]]]
+        ] = bucket_series(
+                                                values = rand_nums,
+                                                outlier_threshold_percent = 10,
+                                                level_granularity=level_granularity
+                                            )
+        
+        bucketized = []
+        for num in rand_nums:
+            bucketized.append(
+                bucketize_val(num, buckets=buckets)
+            )
+
+        
+        range_min = -1
+        range_max = 1
+        size : int = 100
+        percentage_in_range = 91
+        abs_min = -1.5
+        abs_max = 1.5
+
+        rand_nums : List[float] = generate_rand_nums(
+                                        range_min = range_min,
+                                        range_max = range_max,
+                                        size = size, 
+                                        percent_in_range = percentage_in_range,
+                                        abs_min = abs_min,
+                                        abs_max = abs_max
+                                    )
+
+        buckets = bucket_series(
+                                    values = rand_nums,
+                                    outlier_threshold_percent = 10,
+                                    level_granularity=level_granularity
+                                )
+        
+
+        range_min = 0
+        range_max = 100
+        size : int = 100
+        percentage_in_range = 91
+        abs_min = -0.5
+        abs_max = 150
+
+        rand_nums : List[float] = generate_rand_nums(
+                                        range_min = range_min,
+                                        range_max = range_max,
+                                        size = size, 
+                                        percent_in_range = percentage_in_range,
+                                        abs_min = abs_min,
+                                        abs_max = abs_max
+                                    )
+
+        buckets = bucket_series(
+                                    values = rand_nums,
+                                    outlier_threshold_percent = 10,
+                                    level_granularity=level_granularity
+                                )
+        
+
+        range_min = -100
+        range_max = 100
+        size : int = 100
+        percentage_in_range = 91
+        abs_min = -150
+        abs_max = 150
+
+        rand_nums : List[float] = generate_rand_nums(
+                                        range_min = range_min,
+                                        range_max = range_max,
+                                        size = size, 
+                                        percent_in_range = percentage_in_range,
+                                        abs_min = abs_min,
+                                        abs_max = abs_max
+                                    )
+
+        buckets = bucket_series(
+                                    values = rand_nums,
+                                    outlier_threshold_percent = 10,
+                                    level_granularity=level_granularity
+                                )
+        
+
+        range_min = 20_000
+        range_max = 120_000
+        size : int = 100
+        percentage_in_range = 91
+        abs_min = 15_000
+        abs_max = 130_000
+
+        rand_nums : List[float] = generate_rand_nums(
+                                        range_min = range_min,
+                                        range_max = range_max,
+                                        size = size, 
+                                        percent_in_range = percentage_in_range,
+                                        abs_min = abs_min,
+                                        abs_max = abs_max
+                                    )
+
+        buckets = bucket_series(
+                                    values = rand_nums,
+                                    outlier_threshold_percent = 10,
+                                    level_granularity=level_granularity
+                                )

siglab_py/tests/unit/trading_util_tests.py

@@ -0,0 +1,65 @@
+import unittest
+from datetime import datetime, timedelta
+from typing import Union
+from pathlib import Path
+
+from util.trading_util import *
+
+'''
+Have a look at this for a visual explaination how "Gradually tightened stops" works:
+    https://github.com/r0bbar/siglab/blob/master/siglab_py/tests/manual/trading_util_tests.ipynb
+    https://norman-lm-fung.medium.com/gradually-tightened-trailing-stops-f7854bf1e02b
+'''
+
+# @unittest.skip("Skip all integration tests.")
+class TradingUtilTests(unittest.TestCase):
+    def test_calc_eff_trailing_sl_case1(self):
+        tp_min_percent : float = 1.5
+        tp_max_percent : float = 2.5
+        sl_percent_trailing : float = 50 # Trailing stop loss in percent
+        default_effective_tp_trailing_percent : float = 50
+
+        pnl_percent_notional : float = 0.5 # Trade's current pnl in percent.
+
+        effective_tp_trailing_percent = calc_eff_trailing_sl(
+            tp_min_percent = tp_min_percent,
+            tp_max_percent = tp_max_percent,
+            sl_percent_trailing = sl_percent_trailing,
+            pnl_percent_notional = pnl_percent_notional,
+            default_effective_tp_trailing_percent = default_effective_tp_trailing_percent
+        )
+        assert(effective_tp_trailing_percent==50) # Generous trailing SL when trading starting out and pnl small.
+
+    def test_calc_eff_trailing_sl_case2(self):
+        tp_min_percent : float = 1.5
+        tp_max_percent : float = 2.5
+        sl_percent_trailing : float = 50 # Trailing stop loss in percent
+        default_effective_tp_trailing_percent : float = 50
+
+        pnl_percent_notional : float = 2 # Trade's current pnl in percent.
+
+        effective_tp_trailing_percent = calc_eff_trailing_sl(
+            tp_min_percent = tp_min_percent,
+            tp_max_percent = tp_max_percent,
+            sl_percent_trailing = sl_percent_trailing,
+            pnl_percent_notional = pnl_percent_notional,
+            default_effective_tp_trailing_percent = default_effective_tp_trailing_percent
+        )
+        assert(effective_tp_trailing_percent==25) # Intermediate trailing SL
+
+    def test_calc_eff_trailing_sl_case3(self):
+        tp_min_percent : float = 1.5
+        tp_max_percent : float = 2.5
+        sl_percent_trailing : float = 50 # Trailing stop loss in percent
+        default_effective_tp_trailing_percent : float = 50
+
+        pnl_percent_notional : float = 2.5 # Trade's current pnl in percent.
+
+        effective_tp_trailing_percent = calc_eff_trailing_sl(
+            tp_min_percent = tp_min_percent,
+            tp_max_percent = tp_max_percent,
+            sl_percent_trailing = sl_percent_trailing,
+            pnl_percent_notional = pnl_percent_notional,
+            default_effective_tp_trailing_percent = default_effective_tp_trailing_percent
+        )
+        assert(effective_tp_trailing_percent==0) # Most tight trailing SL