keras-team/keras · error · ValueError
Saved configuration not understood. Configuration should be
Error message
Saved configuration not understood. Configuration should be a dictionary, string, tuple or list. Received: config={config}. What it means
Error "Saved configuration not understood. Configuration should be a dictionary, string, tuple or list. Received: config={config}." thrown in keras-team/keras.
Source
Thrown at keras/src/legacy/saving/saving_utils.py:219
if isinstance(obj, str):
return True # Serialized function or string.
return False
if config is None:
return None
if _is_single_object(config):
return deserialize_fn(config)
elif isinstance(config, dict):
return {
k: _deserialize_nested_config(deserialize_fn, v)
for k, v in config.items()
}
elif isinstance(config, (tuple, list)):
return [
_deserialize_nested_config(deserialize_fn, obj) for obj in config
]
raise ValueError(
"Saved configuration not understood. Configuration should be a "
f"dictionary, string, tuple or list. Received: config={config}."
)
def _deserialize_metric(metric_config):
"""Deserialize metrics, leaving special strings untouched."""
if metric_config in ["accuracy", "acc", "crossentropy", "ce"]:
# Do not deserialize accuracy and cross-entropy strings as we have
# special case handling for these in compile, based on model output
# shape.
return metric_config
return metrics_module.deserialize(metric_config)
def _resolve_compile_arguments_compat(obj, obj_config, module):
"""Resolves backwards compatibility issues with training config arguments.
View on GitHub (pinned to 7a34a03db6)
When it happens
Trigger: Thrown at keras/src/legacy/saving/saving_utils.py:219 when the library encounters an invalid state.
Common situations: See trigger scenarios.
AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25).
Data as JSON: /api/errors/65b9f3550e8d44a8.
Report an issue: GitHub.