AttributeError is raised when using RDB storage.

Reproducible examples (optional)

import optuna

# Define a simple 2-dimensional objective function whose minimum value is -1 when (x, y) = (0, -1).
def objective(trial):
    x = trial.suggest_uniform('x', -100, 100)
    y = trial.suggest_categorical('y', [-1, 0, 1])
    return x**2 + y


if __name__ == '__main__':
    # Let us minimize the objective function above.
    print('Running 10 trials...')
    study = optuna.create_study(storage="sqlite:///db.sqlite3")
    study.optimize(objective, n_trials=10)
    print('Best value: {} (params: {})\n'.format(study.best_value, study.best_params))

Expected behavior

tag: v1.1.0

$ git co v1.1.0
$ python examples/quadratic_simple.py 
Running 10 trials...
[I 2020-02-18 14:05:57,875] A new study created with name: no-name-a2889f9d-79d7-40cc-aff4-f4021183b6aa
[I 2020-02-18 14:05:57,977] Finished trial#0 resulted in value: 78.25382960871015. Current best value is 78.25382960871015 with parameters: {'x': -8.78941577175128, 'y': 1}.
[I 2020-02-18 14:05:58,067] Finished trial#1 resulted in value: 1601.2080527969047. Current best value is 78.25382960871015 with parameters: {'x': -8.78941577175128, 'y': 1}.
[I 2020-02-18 14:05:58,159] Finished trial#2 resulted in value: 155.3355136572432. Current best value is 78.25382960871015 with parameters: {'x': -8.78941577175128, 'y': 1}.
[I 2020-02-18 14:05:58,252] Finished trial#3 resulted in value: 4928.445801453334. Current best value is 78.25382960871015 with parameters: {'x': -8.78941577175128, 'y': 1}.
[I 2020-02-18 14:05:58,349] Finished trial#4 resulted in value: 5494.046052713425. Current best value is 78.25382960871015 with parameters: {'x': -8.78941577175128, 'y': 1}.
[I 2020-02-18 14:05:58,451] Finished trial#5 resulted in value: 7.806334896764313. Current best value is 7.806334896764313 with parameters: {'x': -2.793981907021646, 'y': 0}.
[I 2020-02-18 14:05:58,540] Finished trial#6 resulted in value: 3581.7294062867. Current best value is 7.806334896764313 with parameters: {'x': -2.793981907021646, 'y': 0}.
[I 2020-02-18 14:05:58,633] Finished trial#7 resulted in value: 6157.892248662727. Current best value is 7.806334896764313 with parameters: {'x': -2.793981907021646, 'y': 0}.
[I 2020-02-18 14:05:58,729] Finished trial#8 resulted in value: 538.9770074142994. Current best value is 7.806334896764313 with parameters: {'x': -2.793981907021646, 'y': 0}.
[I 2020-02-18 14:05:58,824] Finished trial#9 resulted in value: 2453.3913002520094. Current best value is 7.806334896764313 with parameters: {'x': -2.793981907021646, 'y': 0}.
Best value: 7.806334896764313 (params: {'x': -2.793981907021646, 'y': 0})

Error log

rev: 7344ee0f

$ git co master
$ rm db.sqlite3 | true && python examples/quadratic_simple.py 
Running 10 trials...
[I 2020-02-18 14:06:51,506] A new study created with name: no-name-32a1bb2f-64c7-4faa-8834-0bbdbaff508b
[I 2020-02-18 14:06:51,635] Finished trial#0 resulted in value: 2420.888780807176. Current best value is 2420.888780807176 with parameters: {'x': -49.192365066208964, 'y': 1}.
[I 2020-02-18 14:06:51,746] Finished trial#1 resulted in value: 145.24171551709355. Current best value is 145.24171551709355 with parameters: {'x': 12.093044096384233, 'y': -1}.
[I 2020-02-18 14:06:51,857] Finished trial#2 resulted in value: 1.014598248626613. Current best value is 1.014598248626613 with parameters: {'x': -0.12082321228395188, 'y': 1}.
[I 2020-02-18 14:06:51,957] Finished trial#3 resulted in value: 8577.24930691218. Current best value is 1.014598248626613 with parameters: {'x': -0.12082321228395188, 'y': 1}.
[I 2020-02-18 14:06:52,068] Finished trial#4 resulted in value: 8417.027436775961. Current best value is 1.014598248626613 with parameters: {'x': -0.12082321228395188, 'y': 1}.
[I 2020-02-18 14:06:52,181] Finished trial#5 resulted in value: 319.0475577707139. Current best value is 1.014598248626613 with parameters: {'x': -0.12082321228395188, 'y': 1}.
[I 2020-02-18 14:06:52,288] Finished trial#6 resulted in value: 4.4949007058993296. Current best value is 1.014598248626613 with parameters: {'x': -0.12082321228395188, 'y': 1}.
[I 2020-02-18 14:06:52,391] Finished trial#7 resulted in value: 2865.4595762936938. Current best value is 1.014598248626613 with parameters: {'x': -0.12082321228395188, 'y': 1}.
[I 2020-02-18 14:06:52,500] Finished trial#8 resulted in value: 2027.0173870485705. Current best value is 1.014598248626613 with parameters: {'x': -0.12082321228395188, 'y': 1}.
[I 2020-02-18 14:06:52,605] Finished trial#9 resulted in value: 7283.744078340451. Current best value is 1.014598248626613 with parameters: {'x': -0.12082321228395188, 'y': 1}.
Best value: 1.014598248626613 (params: {'x': -0.12082321228395188, 'y': 1})

Exception ignored in: <function RDBStorage.__del__ at 0x130bd7ca0>
Traceback (most recent call last):
  File "/Users/a14737/src/github.com/pfnet/optuna/optuna/storages/rdb/storage.py", line 943, in __del__
  File "/Users/a14737/src/github.com/pfnet/optuna/optuna/storages/rdb/storage.py", line 932, in remove_session
  File "/Users/a14737/src/github.com/pfnet/optuna/venv/lib/python3.8/site-packages/SQLAlchemy-1.3.12-py3.8-macosx-10.15-x86_64.egg/sqlalchemy/orm/scoping.py", line 94, in remove
  File "/Users/a14737/src/github.com/pfnet/optuna/venv/lib/python3.8/site-packages/SQLAlchemy-1.3.12-py3.8-macosx-10.15-x86_64.egg/sqlalchemy/orm/session.py", line 1298, in close
  File "/Users/a14737/src/github.com/pfnet/optuna/venv/lib/python3.8/site-packages/SQLAlchemy-1.3.12-py3.8-macosx-10.15-x86_64.egg/sqlalchemy/orm/session.py", line 1334, in _close_impl
  File "/Users/a14737/src/github.com/pfnet/optuna/venv/lib/python3.8/site-packages/SQLAlchemy-1.3.12-py3.8-macosx-10.15-x86_64.egg/sqlalchemy/orm/session.py", line 1352, in expunge_all
AttributeError: 'NoneType' object has no attribute 'InstanceState'

Environment

• Optuna version: 1.2.0 (master branch, rev: 7344ee0f).
• Python version: 3.8.1
• OS: macOS
• (Optional) Other libraries and their versions:

Steps to reproduce

1. Optimize with RDB storage backend.

Additional context (optional)

[toshihikoyanase]

I failed to reproduce this problem when I used docker as follows:

938.py

import optuna

# Define a simple 2-dimensional objective function whose minimum value is -1 when (x, y) = (0, -1).
def objective(trial):
    x = trial.suggest_uniform('x', -100, 100)
    y = trial.suggest_categorical('y', [-1, 0, 1])
    return x**2 + y


if __name__ == '__main__':
    # Let us minimize the objective function above.
    print('Running 10 trials...')
    study = optuna.create_study(storage="sqlite:///db.sqlite3")
    study.optimize(objective, n_trials=10)
    print('Best value: {} (params: {})\n'.format(study.best_value, study.best_params))

$ docker run -v $(pwd)/:/var/examples --rm -it python:3.8.1 /bin/bash

# pip install git+https://github.com/optuna/optuna.git@7344ee0fcdddb28b18a292e63caea425ae08da95
# python 938.py

Could you share more information about your environment such as the output of pip freeze?

[c-bata]

I once closed this but I wonder that someone will report the same issue because prompt_toolkit is in dependencies of jupyter. Hmm…

(venv) $ pip install jupyter

...

Successfully installed MarkupSafe-1.1.1 Send2Trash-1.5.0 appnope-0.1.0 attrs-19.3.0 backcall-0.1.0 bleach-3.1.0 decorator-4.4.1 defusedxml-0.6.0 entrypoints-0.3 ipykernel-5.1.4 ipython-7.12.0 ipython-genutils-0.2.0 ipywidgets-7.5.1 jedi-0.16.0 jinja2-2.11.1 jsonschema-3.2.0 jupyter-1.0.0 jupyter-client-5.3.4 jupyter-console-6.1.0 jupyter-core-4.6.2 mistune-0.8.4 nbconvert-5.6.1 nbformat-5.0.4 notebook-6.0.3 pandocfilters-1.4.2 parso-0.6.1 pexpect-4.8.0 pickleshare-0.7.5 prometheus-client-0.7.1 prompt-toolkit-3.0.3 ptyprocess-0.6.0 pygments-2.5.2 pyrsistent-0.15.7 python-dateutil-2.8.1 pyzmq-18.1.1 qtconsole-4.6.0 six-1.14.0 terminado-0.8.3 testpath-0.4.4 tornado-6.0.3 traitlets-4.3.3 wcwidth-0.1.8 webencodings-0.5.1 widgetsnbextension-3.5.1