renaming file

fast_trade/__init__.py

@@ -5,4 +5,4 @@
 from .transformers_map import transformers_map
 from .validate_backtest import validate_backtest
 from .update_symbol_data import update_symbol_data, load_archive_to_df
-from .check_missing_dates import check_missing_dates
+from .calculate_perc_missing import calculate_perc_missing

fast_trade/build_summary.py

@@ -2,6 +2,7 @@
 
 import numpy as np
 import pandas as pd
+from .calculate_perc_missing import calculate_perc_missing
 
 
 def build_summary(df, performance_start_time):
@@ -58,6 +59,8 @@ def build_summary(df, performance_start_time):
 
     performance_stop_time = datetime.datetime.utcnow()
 
+    perc_missing = calculate_perc_missing(df)
+
     summary = {
         "return_perc": return_perc,
         "sharpe_ratio": sharpe_ratio,  # BETA
@@ -84,6 +87,7 @@ def build_summary(df, performance_start_time):
         "first_tic": start_date.strftime("%Y-%m-%d %H:%M:%S"),
         "last_tic": end_date.strftime("%Y-%m-%d %H:%M:%S"),
         "total_tics": len(df.index),
+        "perc_missing": perc_missing,
         "test_duration": round(
             (performance_stop_time - performance_start_time).total_seconds(), 3
         ),

fast_trade/calculate_perc_missing.py

@@ -0,0 +1,22 @@
+def calculate_perc_missing(df):
+    """
+    Parameters
+    ----------
+        df, pandas dataframe check
+
+    Returns
+    -------
+        perc_missing, float, a float of the percentage of rows with the missing closing price
+
+    """
+
+    total_missing = df.close.isna().sum()
+
+    perc_missing = 0.0
+    if total_missing:
+        total_tics = len(df.index)
+        perc_missing = (total_missing / total_tics) * 100
+
+        perc_missing = round(perc_missing, 2)
+
+    return perc_missing, total_missing

fast_trade/cli.py

@@ -105,12 +105,12 @@ def main():
         return
 
     if command == "validate":
-        print("args: ",args)
+        print("args: ", args)
         backtest = open_strat_file(args["backtest"])
         if not backtest:
             print("backtest not found! ")
             return
-        print("backtest: ",backtest)
+        print("backtest: ", backtest)
         backtest = {**backtest, **args}
 
         res = validate_backtest(backtest)

fast_trade/run_analysis.py

@@ -73,7 +73,6 @@ def apply_logic_to_df(df: pd.DataFrame, backtest: dict):
         adj_account_value = new_account_value + convert_aux_to_base(aux, close)
 
         aux_list.append(aux)
-        # base_list.append(new_base)
         account_value_list.append(new_account_value)
         in_trade_list.append(in_trade)
         fee_list.append(fee)
@@ -99,7 +98,6 @@ def apply_logic_to_df(df: pd.DataFrame, backtest: dict):
         adj_account_value_list.append(adj_account_value)
 
     df["aux"] = aux_list
-    # df["base"] = base_list
     df["account_value"] = account_value_list
     df["adj_account_value"] = adj_account_value_list
     df["in_trade"] = in_trade_list

fast_trade/run_backtest.py

@@ -45,7 +45,6 @@ def run_backtest(
         "summary": summary,
         "df": df,
         "trade_df": trade_log,
-        # "missing_data": gaps,
         "backtest": new_backtest,
     }
 

run_example.py

@@ -2,7 +2,7 @@
 from fast_trade.build_data_frame import prepare_df
 from fast_trade import run_backtest
 from fast_trade.validate_backtest import validate_backtest
-from fast_trade.check_missing_dates import check_missing_dates
+from fast_trade.calculate_perc_missing import calculate_perc_missing
 import datetime
 import json
 import pandas as pd
@@ -27,56 +27,55 @@
 # case 1: sell
 
 # case 2: buy
-backtest = {
-    "any_enter": [],
-    "any_exit": [],
-    "chart_period": "1H",
-    "comission": 0.00,
-    "base_balance": 100000,
-    "lot_size": 0.6,
-    "datapoints": [{"args": [], "name": "rsi_dp", "transformer": "rsi"}],
-    "enter": [
-        [
-            "rsi_dp",
-            "<",
-            30,
-        ]
-    ],
-    "exit": [
-        [
-            "rsi_dp",
-            ">",
-            70,
-        ]
-    ],
-    "exit_on_end": False,
-    "max_lot_size": 0
-    # "start": "2019-12-01 15:29:00",
-    # "stop": "2021-03-08 15:29:00",
-    # "trailing_stop_loss": 0.05,  # 5% stoploss
-}
-
 # backtest = {
 #     "any_enter": [],
 #     "any_exit": [],
-#     # "chart_period": "37Min",
-#     "chart_period": "11Min",
+#     "chart_period": "15Min",
 #     "comission": 0.01,
-#     "datapoints": [
-#         {"args": [14], "name": "er", "transformer": "er"},
-#         {"args": [14], "name": "zlema_1", "transformer": "zlema"},
-#         {"args": [66], "name": "zlema_2", "transformer": "zlema"},
+#     "base_balance": 100000,
+#     "lot_size": 0.6,
+#     "datapoints": [{"args": [], "name": "rsi_dp", "transformer": "rsi"}],
+#     "enter": [
+#         [
+#             "rsi_dp",
+#             "<",
+#             30,
+#         ]
+#     ],
+#     "exit": [
+#         [
+#             "rsi_dp",
+#             ">",
+#             70,
+#         ]
 #     ],
-#     "enter": [["zlema_1", ">", "close", 1]],
-#     "exit": [["zlema_2", "<", "close", 1]],
 #     "exit_on_end": False,
-#     # "start": "2021-01-01 22:30:00",
-#     # "stop": "2021-03-11 23:30:59",
-#     "trailing_stop_loss": 0.05,
-#     # "max_lot_size": 1000,
-#     "lot_size": 1,
-#     "base_balance": 500,
+#     "max_lot_size": 0
+#     # "start": "2019-12-01 15:29:00",
+#     # "stop": "2021-03-08 15:29:00",
+#     # "trailing_stop_loss": 0.05,  # 5% stoploss
 # }
+
+backtest = {
+    "any_enter": [],
+    "any_exit": [],
+    # "chart_period": "37Min",
+    "chart_period": "1H",
+    "comission": 0.01,
+    "datapoints": [
+        {"args": [11], "name": "zlema_1", "transformer": "zlema"},
+        {"args": [88], "name": "zlema_2", "transformer": "zlema"},
+    ],
+    "enter": [["zlema_1", ">", "close", 4]],
+    "exit": [["zlema_2", "<", "close", 1]],
+    "exit_on_end": False,
+    # "start": "2021-01-01 22:30:00",
+    # "stop": "2021-03-11 23:30:59",
+    # "trailing_stop_loss": 0.05,
+    # "max_lot_size": 1000,
+    "lot_size": 1,
+    "base_balance": 500,
+}
 # backtest = {
 #     "any_enter": [],
 #     "any_exit": [],
@@ -93,25 +92,19 @@
 if __name__ == "__main__":
     # datafile = "./BTCUSDT.csv"
     # datafile = "./archive/BTCUSDT_2021.csv"
-    datafile = "/Users/jedmeier/Desktop/BTCUSDT_ALL/BTCUSDT_2021.csv"
+    datafile = "/Users/jedmeier/Desktop/BTCUSDT_ALL/BTCUSDT_2020.csv"
     # datafile = "/Users/jedmeier/Desktop/BTCUSDT_2021_06_12.csv"
     # datafile = "./archive/BCHUSDT_2021.csv"
 
-    # df = pd.read_csv(datafile)
-
-    # df.index = pd.to_datetime(df.date, unit="s")
-    # print(df)
-    # check_missing_dates(df)
-    # with open("./example_backtest.json", "r") as backtest_file:
-    #     backtest = json.load(backtest_file)
-    # print(backtest)
     # backtest["start"] = "2021-03-01"
     # backtest["stop"] = "2021-05-01"
 
     # backtest["chart_period"] = "1Min"
     test = run_backtest(backtest, datafile, summary=True)
     # print(test)
-    df = test["trade_df"]
+    df = test["df"]
+
+    perc_missing = calculate_perc_missing(df)
 
-    print(df)
-    print(json.dumps(test["summary"], indent=2))
+    print("perc_missing: ", perc_missing)
+    # print(json.dumps(test["summary"], indent=2))

setup.py

@@ -7,7 +7,7 @@
 
 setup(
     name="fast-trade",
-    version="0.2.3",
+    version="0.3.0",
     description="About low code backtesting library utilizing pandas and technical analysis indicators",
     long_description=README,
     long_description_content_type="text/markdown",

test/test_calculate_perc_missing.py

@@ -0,0 +1,35 @@
+from datetime import datetime
+import pandas as pd
+import numpy as np
+from pandas.core.base import DataError
+from fast_trade.calculate_perc_missing import calculate_perc_missing
+
+
+def test_calculate_perc_missing_none_missing():
+    mock_df = pd.read_csv("./test/ohlcv_data.csv.txt", parse_dates=True).set_index(
+        "date"
+    )
+
+    [perc_missinng, total_missing] = calculate_perc_missing(mock_df)
+
+    assert perc_missinng == 0.0
+    assert total_missing == 0
+
+
+def test_calculate_perc_missing_some_missing():
+    mock_df = pd.read_csv("./test/ohlcv_data.csv.txt", parse_dates=True)
+    # mock_df.index = pd.to_datetime(mock_df.date, unit="s")
+    # print(mock_df.iloc[0].name)
+    print(mock_df.iloc[0].date)
+    # wtf = mock_df.index.get_loc(pd.to_datetime(1523938263, unit="s"))
+
+    # print(mock_df.iloc[wtf])
+    mock_df.close = [np.nan, np.nan, 2, 3, 4, 5, 6, 7, 8]
+    # print(mock_df["1523937784"])
+    # mock_df[1523937784].close = np.nan
+    [perc_missinng, total_missing] = calculate_perc_missing(mock_df)
+    #
+    assert perc_missinng == 22.22
+    assert total_missing == 2
+    # print(perc_missinng)
+    # assert True is False