Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

suggest_discrete_uniform error "unsupported operand type(s) for *: 'int' and 'NoneType'"

See original GitHub issue

Conditions

Optuna version: 0.14.0
Python version: 3.6.8
OS: Google Colab
Machine Learning library to be optimized: TensorFlow 2.0.0-beta1

Code to reproduce

The error started during a study run on Google Colab. I don’t have a min repro but can share the model sqlite DB and model code if needed.

Error messages, stack traces, or logs

[W 2019-08-05 19:12:03,795] Setting status of trial#27 as TrialState.FAIL because of the following error: TypeError("unsupported operand type(s) for *: 'int' and 'NoneType'",)
Traceback (most recent call last):
  File "/usr/local/lib/python3.6/dist-packages/optuna/study.py", line 469, in _run_trial
    result = func(trial)
  File "<ipython-input-42-176311fa1b08>", line 33, in objective
    model = make_trial_model(trial, input_shape)
  File "<ipython-input-17-3a5c1667bd48>", line 37, in make_trial_model
    embedding_size = trial.suggest_discrete_uniform('embedding_size', 100, 300, 50)
  File "/usr/local/lib/python3.6/dist-packages/optuna/trial.py", line 268, in suggest_discrete_uniform
    return self._suggest(name, distribution)
  File "/usr/local/lib/python3.6/dist-packages/optuna/trial.py", line 455, in _suggest
    study, trial, name, distribution)
  File "/usr/local/lib/python3.6/dist-packages/optuna/samplers/tpe/sampler.py", line 94, in sample_independent
    values, scores = _get_observation_pairs(study, param_name)
  File "/usr/local/lib/python3.6/dist-packages/optuna/samplers/tpe/sampler.py", line 523, in _get_observation_pairs
    score = (-step, sign * intermediate_value)
TypeError: unsupported operand type(s) for *: 'int' and 'NoneType'