microsoft/qlib · error · ValueError

report_type should be one of pred_long, pred_long_short, pre

Error message

report_type should be one of pred_long, pred_long_short, pred_short, excess_return_without_cost, excess_return_with_cost and model

What it means

OptimizationConfig (qlib/contrib/tuner/config.py) validates the report_type field of the tuner YAML. Only six report types are accepted: pred_long, pred_long_short, pred_short, excess_return_without_cost, excess_return_with_cost, and model. Anything else raises ValueError at config load time.

Source

Thrown at qlib/contrib/tuner/config.py:70

        self.tuner_class = config.get("tuner_class", "QLibTuner")
        # Save the tuner experiment for further view
        tuner_ex_config_path = os.path.join(self.tuner_ex_dir, "tuner_config.yaml")
        with open(tuner_ex_config_path, "w") as fp:
            yaml.dump(TUNER_CONFIG_MANAGER.config, fp)


class OptimizationConfig:
    def __init__(self, config, TUNER_CONFIG_MANAGER):
        self.report_type = config.get("report_type", "pred_long")
        if self.report_type not in [
            "pred_long",
            "pred_long_short",
            "pred_short",
            "excess_return_without_cost",
            "excess_return_with_cost",
            "model",
        ]:
            raise ValueError(
                "report_type should be one of pred_long, pred_long_short, pred_short, excess_return_without_cost, excess_return_with_cost and model"
            )

        self.report_factor = config.get("report_factor", "information_ratio")
        if self.report_factor not in [
            "annualized_return",
            "information_ratio",
            "max_drawdown",
            "mean",
            "std",
            "model_score",
            "model_pearsonr",
        ]:
            raise ValueError(
                "report_factor should be one of annualized_return, information_ratio, max_drawdown, mean, std, model_pearsonr and model_score"
            )

        self.optim_type = config.get("optim_type", "max")

View on GitHub (pinned to 79633dd950)

Solutions

1. Set report_type to one of: pred_long, pred_long_short, pred_short, excess_return_without_cost, excess_return_with_cost, model
2. Check spelling and case in optimization_criteria.report_type in your YAML
3. Upgrade/downgrade qlib to the version whose tuner examples you are following if the accepted set differs

Example fix

# before (tuner_config.yaml)
optimization_criteria:
  report_type: long

# after
optimization_criteria:
  report_type: pred_long

VALID_REPORT_TYPES = {'pred_long', 'pred_long_short', 'pred_short', 'excess_return_without_cost', 'excess_return_with_cost', 'model'}
assert cfg['optimization_criteria']['report_type'] in VALID_REPORT_TYPES

def is_valid_report_type(rt: str) -> bool:
    return rt in {'pred_long', 'pred_long_short', 'pred_short', 'excess_return_without_cost', 'excess_return_with_cost', 'model'}

When it happens

Trigger: A tuner YAML with optimization_criteria: report_type: long (or any unsupported/misspelled/case-different value) is loaded by TunerConfigManager; the error fires during OptimizationConfig construction.

Common situations: Hand-writing the tuner config and abbreviating the report type; version drift where older/newer qlib accepts a different set of report types; copy-pasting a report type from analysis-report code (e.g. 'report_long') that is not valid here.

Related errors

• Config path is invalid.
• report_factor should be one of annualized_return, informatio
• optim_type should be min, max or correlation
• Please give the search space of model.
• Please give the search space of strategy.

AI-assisted analysis of microsoft/qlib@79633dd950 (2026-08-15). Data as JSON: /api/errors/c140edeccd424e51. Report an issue: GitHub.