keras-team/keras

[Bug] deserialize_keras_object raises AttributeError for invalid config types

Open

#22701 opened on Apr 18, 2026

View on GitHub
 (6 comments) (0 reactions) (1 assignee)Python (64,067 stars) (19,774 forks)batch import
Good first issuebackend:tensorflowstat:contributions welcometype:Bug

Description

Description

deserialize_keras_object may raise AttributeError when provided with invalid config types (e.g., list or string), indicating that input validation occurs after internal method calls.

Reproduction Script

import os
os.environ["KERAS_BACKEND"] = "tensorflow"

from keras.saving import deserialize_keras_object

config = {
    "class_name": "Dense",
    "module": "keras.layers",
    "config": [1, 2, 3]   # invalid type
}

try:
    deserialize_keras_object(config)
except Exception as e:
    print("Exception type:", type(e).__name__)
    print("Message:", e)

Observed behavior

  • Raises AttributeError: 'list' object has no attribute 'get'

Expected behavior

  • Should raise a clear ValueError or TypeError for invalid input structure

Notes

This suggests input validation happens too late and exposes internal implementation details.

Contributor guide

Tech stack
python
Domain
backendmachine learning
Issue type
bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
blocked
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
clear
Prerequisites
PythonError handling
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
30
Research direction
The bug is in the `deserialize keras object` function, likely in `keras/saving/`. The fix involves adding input validation at the start to ensure `config` is a dict and raising a clear `ValueError` or `TypeError` instead of an `AttributeError`. Examine the existing comments in the issue for any suggested approaches or additional context. The change is localized and should include a simple type check before calling `.get()` on the config.