Backtest too slow when using my own data

[TompaBay]

I'm using CSMAR's data for chinese market and use following backtest code:

`from pprint import pprint

import qlib import pandas as pd from qlib.utils.time import Freq from qlib.utils import flatten_dict from qlib.backtest import backtest, executor from qlib.contrib.evaluate import risk_analysis from qlib.contrib.strategy import TopkDropoutStrategy

if name == "main":

qlib.init(provider_uri=r"../../benchmark/cn_data/qlib_data/")
score_df = pd.read_csv("../pred.csv")
score_df["datetime"] = pd.to_datetime(score_df["datetime"])
pred_score = score_df.set_index(["datetime", "instrument"])["score"]
CSI300_BENCH = "SH000300"
FREQ = "day"
STRATEGY_CONFIG = {
    "topk": 50,
    "n_drop": 10,
    "signal": pred_score,

}

EXECUTOR_CONFIG = {
    "time_per_step": "day",
    "generate_portfolio_metrics": True,
    "verbose": True,
}

backtest_config = {
    "start_time": "2016-01-01",
    "end_time": "2016-12-31",
    "account": 100000000,
    "benchmark": CSI300_BENCH,
    "exchange_kwargs": {
        "trade_unit": 100,
        "freq": FREQ,
        "limit_threshold": 0.095,
        "deal_price": "close",
        "open_cost": 0.0015,
        "close_cost": 0.0025,
        "min_cost": 5,
    },
}

strategy_obj = TopkDropoutStrategy(**STRATEGY_CONFIG)
executor_obj = executor.SimulatorExecutor(**EXECUTOR_CONFIG)

portfolio_metric_dict, indicator_dict = backtest(executor=executor_obj, strategy=strategy_obj, **backtest_config)
analysis_freq = "{0}{1}".format(*Freq.parse(FREQ))

report_normal, positions_normal = portfolio_metric_dict.get(analysis_freq)
analysis = dict()
analysis["excess_return_without_cost"] = risk_analysis(
    report_normal["return"] - report_normal["bench"], freq=analysis_freq
)
analysis["excess_return_with_cost"] = risk_analysis(
    report_normal["return"] - report_normal["bench"] - report_normal["cost"], freq=analysis_freq
)

analysis_df = pd.concat(analysis)  # type: pd.DataFrame
# log metrics
analysis_dict = flatten_dict(analysis_df["risk"].unstack().T.to_dict())
# print out results
pprint(f"The following are analysis results of benchmark return({analysis_freq}).")
pprint(risk_analysis(report_normal["bench"], freq=analysis_freq))
pprint(f"The following are analysis results of the excess return without cost({analysis_freq}).")
pprint(analysis["excess_return_without_cost"])
pprint(f"The following are analysis results of the excess return with cost({analysis_freq}).")
pprint(analysis["excess_return_with_cost"])`

The whole process was too slow that it took about 1 hour to test on just 1 year. Wonder what could be the reason? I do see the "future error", "no common_infra" and "nan in close" warning errors.

[TompaBay]

It does show warnings like image above.