mlflow/mlflow · warning · ValueError

The two merging dictionaries contains duplicate keys: {dupli

Error message

The two merging dictionaries contains duplicate keys: {duplicate_keys}.

What it means

mlflow.utils.merge_dicts merges two dicts but, when raise_on_duplicates=True (the default), refuses to silently overwrite keys present in both inputs. It raises ValueError listing the duplicate keys so data loss is explicit.

Source

Thrown at mlflow/utils/__init__.py:132

    return truncated


def merge_dicts(dict_a, dict_b, raise_on_duplicates=True):
    """This function takes two dictionaries and returns one singular merged dictionary.

    Args:
        dict_a: The first dictionary.
        dict_b: The second dictionary.
        raise_on_duplicates: If True, the function raises ValueError if there are duplicate keys.
            Otherwise, duplicate keys in `dict_b` will override the ones in `dict_a`.

    Returns:
        A merged dictionary.

    """
    duplicate_keys = dict_a.keys() & dict_b.keys()
    if raise_on_duplicates and len(duplicate_keys) > 0:
        raise ValueError(f"The two merging dictionaries contains duplicate keys: {duplicate_keys}.")
    return dict_a | dict_b


def _get_fully_qualified_class_name(obj):
    """
    Obtains the fully qualified class name of the given object.
    """
    return obj.__class__.__module__ + "." + obj.__class__.__name__


def _inspect_original_var_name(var, fallback_name):
    """
    Inspect variable name, will search above frames and fetch the same instance variable name
    in the most outer frame.
    If inspect failed, return fallback_name
    """
    if var is None:
        return fallback_name

View on GitHub (pinned to 6a27f2decc)

Solutions

1. Remove the duplicate keys from one dict before merging
2. Pass raise_on_duplicates=False if last-wins (dict_b) overwrite is acceptable
3. Rename keys to make them unique before merging

Example fix

// before
merged = merge_dicts(defaults, user_cfg)  # KeyError-free but ValueError on dup keys
// after
merged = merge_dicts(defaults, user_cfg, raise_on_duplicates=False)  # or de-duplicate user_cfg first

dups = dict_a.keys() & dict_b.keys()
if dups:
    print(f"Duplicate keys will raise: {dups}")

try:
    merged = merge_dicts(dict_a, dict_b)
except ValueError as e:
    # decide policy: last-wins merge
    merged = {**dict_a, **dict_b}

When it happens

Trigger: Calling merge_dicts(dict_a, dict_b) where the dicts share one or more keys, e.g. merging two params/config dicts built from overlapping sources; also directly in tests.

Common situations: Merging default and user config that share keys; combining run params from two runs with overlapping names; environment-derived dicts colliding with defaults.

Related errors

• Must specify at least either `max_key_length` or `max_value_
• Cannot resolve {raw_source} as an evaluation dataset ID
• Template must be a list of dicts with role and content
• Use _BooleanEnvironmentVariable instead for boolean variable
• Failed to convert {val!r} for {self.name}: {e}

AI-assisted analysis of mlflow/mlflow@6a27f2decc (2026-08-29). Data as JSON: /api/errors/2c300568b30c6660. Report an issue: GitHub.