Exception: The ray.global_state API cannot be used before ray.init has been called.

System information

• OS Platform and Distribution (e.g., Linux Ubuntu 16.04): CentOS Linux release 7.5.1804 (Core)
• Modin installed from (source or binary): pip install modin
• Modin version: 0.2.5
• Python version: 3.5.2
• Ray version: 0.6.0
• Pandas version: 0.23.4
• Exact command to reproduce: python manage.py runserver with a view file starting by import modin.pandas as pd

Describe the problem

When I do python -c 'import modin.pandas as pd' everything works fine. But when I try to start a django server containing the same import, I get the following error :

Exception: The ray.global_state API cannot be used before ray.init has been called.

Is modin designed to work with Django ?

Source code / logs

 urlconf_module = import_module(urlconf_module)
  File "/home/.pyenv/versions/3.5.2/lib/python3.5/importlib/__init__.py", line 126, in import_module
    return _bootstrap._gcd_import(name[level:], package, level)
  File "<frozen importlib._bootstrap>", line 986, in _gcd_import
  File "<frozen importlib._bootstrap>", line 969, in _find_and_load
  File "<frozen importlib._bootstrap>", line 958, in _find_and_load_unlocked
  File "<frozen importlib._bootstrap>", line 673, in _load_unlocked
  File "<frozen importlib._bootstrap_external>", line 665, in exec_module
  File "<frozen importlib._bootstrap>", line 222, in _call_with_frames_removed
  File "/home/Projects/NoE/core/urls.py", line 3, in <module>
    from core.views import (barcodes, classifications, network_samples, provider,
  File "/home/Projects/NoE/core/views/sample_file_load.py", line 1, in <module>
    import modin.pandas as pd
  File "/home/.pyenv/versions/python352/lib/python3.5/site-packages/modin/pandas/__init__.py", line 95, in <module>
    num_cpus = ray.global_state.cluster_resources()["CPU"]
  File "/home/.pyenv/versions/python352/lib/python3.5/site-packages/ray/experimental/state.py", line 754, in cluster_resources
    clients = self.client_table()
  File "/home/.pyenv/versions/python352/lib/python3.5/site-packages/ray/experimental/state.py", line 321, in client_table
    self._check_connected()
  File "/home/.pyenv/versions/python352/lib/python3.5/site-packages/ray/experimental/state.py", line 45, in _check_connected
    raise Exception("The ray.global_state API cannot be used before "
Exception: The ray.global_state API cannot be used before ray.init has been called.

[Clem-D]

Hello, Actually it works ! It also works when I do :

import ray
ray.init()
ray.global_state.cluster_resources()["CPU"]
import modin.pandas as pd

Now the problem comes from read_excel function… I used to use " encoding=‘latin1’ " which seems to be an unknown keyword with modin but works with panda. I mean with the above import lines I get the following error :

read_excel() got an unexpected keyword argument 'encoding'

But if I replace import modin.pandas as pd by import pandas as pd it works… Should I open an other ticket ?

Thanks for your help.

[devin-petersohn]

Closing this. Feel free to reopen if the discussion should continue or if issue was not resolved.