FoundationAgents/MetaGPT · error · ValueError

Dataset {dataset_name} not found in config file. Available d

Error message

Dataset {dataset_name} not found in config file. Available datasets: {config['datasets'].keys()}

What it means

Raised by the SELA dataset loader when the requested dataset_name is not a key in the loaded config's 'datasets' mapping (the datasets.yaml registry). The message lists the datasets that ARE available so the mismatch is visible.

Source

Thrown at metagpt/ext/sela/data/dataset.py:128

def get_split_dataset_path(dataset_name, config):
    datasets_dir = config["datasets_dir"]
    if dataset_name in config["datasets"]:
        dataset = config["datasets"][dataset_name]
        data_path = os.path.join(datasets_dir, dataset["dataset"])
        split_datasets = {
            "train": os.path.join(data_path, "split_train.csv"),
            "dev": os.path.join(data_path, "split_dev.csv"),
            "dev_wo_target": os.path.join(data_path, "split_dev_wo_target.csv"),
            "dev_target": os.path.join(data_path, "split_dev_target.csv"),
            "test": os.path.join(data_path, "split_test.csv"),
            "test_wo_target": os.path.join(data_path, "split_test_wo_target.csv"),
            "test_target": os.path.join(data_path, "split_test_target.csv"),
        }
        return split_datasets
    else:
        raise ValueError(
            f"Dataset {dataset_name} not found in config file. Available datasets: {config['datasets'].keys()}"
        )


def get_user_requirement(task_name, config):
    # datasets_dir = config["datasets_dir"]
    if task_name in config["datasets"]:
        dataset = config["datasets"][task_name]
        # data_path = os.path.join(datasets_dir, dataset["dataset"])
        user_requirement = dataset["user_requirement"]
        return user_requirement
    else:
        raise ValueError(
            f"Dataset {task_name} not found in config file. Available datasets: {config['datasets'].keys()}"
        )


def save_datasets_dict_to_yaml(datasets_dict, name="datasets.yaml"):

View on GitHub (pinned to 11cdf466d0)

Solutions

1. Use one of the dataset names printed in the error message (they are the available keys)
2. Register your dataset in the datasets config yaml with its 'dataset', 'target_col' and 'user_requirement' entries
3. Check for typos/case differences in the task name passed on the CLI

Example fix

# before (task not registered)
datasets = get_split_datasets("abaline", config)

# after
datasets = get_split_datasets("abalone", config)

if dataset_name not in config["datasets"]:
    raise KeyError(f"unknown dataset {dataset_name}; known: {sorted(config['datasets'])}")

def is_registered_dataset(name: str, config: dict) -> bool:
    return name in config.get("datasets", {})

When it happens

Trigger: Calling the split-dataset helper with a task/dataset name that is not registered, e.g. get_split_datasets('abalone', config) or a custom OpenML dataset that was never added to datasets.yaml.

Common situations: Typos in --task; using a task name from a different fork/branch of datasets.yaml; forgetting to register a newly added dataset in config/datasets.yaml.

Related errors

• Dataset {task_name} not found in config file. Available data
• Target column not provided
• Number of classes {num_classes} not supported
• Invalid exp_mode: {args.exp_mode}
• Dataset {task_name} not found in config file. Available data

AI-assisted analysis of FoundationAgents/MetaGPT@11cdf466d0 (2026-08-14). Data as JSON: /api/errors/4155b55bd27de524. Report an issue: GitHub.