Is training function call included or only the infer function?

Competition Obesity ML Challenge Part 3
,
1

Will there be train(..) function call execution or not. In the baseline notebook there is not mention of train function. Will only infer(…) function be used?

2

There is no train function for this competition. Just an infer function.

If you need to train a model, you must do it in the infer.

3

So, when submitting the new model, I am getting an error after the upload:

`Current Crunch not found`

Is it something related to refresh at every Tuesday? When can I resume uploading my models?

4

Apologies for the issue.

The system responsible to refresh the quota failed.

It is fixed!

5

Getting an error

Traceback (most recent call last):

  File "<frozen runpy>", line 198, in _run_module_as_main

  File "<frozen runpy>", line 88, in _run_code

  File "/context/venv/lib/python3.12/site-packages/crunch/__main__.py", line 3, in <module>

    cli()

  File "/context/venv/lib/python3.12/site-packages/click/core.py", line 1524, in __call__

    return self.main(*args, **kwargs)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/context/venv/lib/python3.12/site-packages/click/core.py", line 1445, in main

    rv = self.invoke(ctx)

         ^^^^^^^^^^^^^^^^

  File "/context/venv/lib/python3.12/site-packages/click/core.py", line 1912, in invoke

    return _process_result(sub_ctx.command.invoke(sub_ctx))

                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/context/venv/lib/python3.12/site-packages/click/core.py", line 1912, in invoke

    return _process_result(sub_ctx.command.invoke(sub_ctx))

                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/context/venv/lib/python3.12/site-packages/click/core.py", line 1308, in invoke

    return ctx.invoke(self.callback, **ctx.params)

           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/context/venv/lib/python3.12/site-packages/click/core.py", line 877, in invoke

    return callback(*args, **kwargs)

           ^^^^^^^^^^^^^^^^^^^^^^^^^

  File "/context/venv/lib/python3.12/site-packages/crunch/cli.py", line 813, in cloud_executor

    executor.start()

  File "/context/venv/lib/python3.12/site-packages/crunch/runner/cloud_executor.py", line 103, in start

    self._process_unstructured()

  File "/context/venv/lib/python3.12/site-packages/crunch/runner/cloud_executor.py", line 151, in _process_unstructured

    smart_call(

  File "/context/venv/lib/python3.12/site-packages/crunch/utils.py", line 233, in smart_call

    return function(**arguments)

           ^^^^^^^^^^^^^^^^^^^^^

  File "/context/scoring/runner.py", line 45, in infer

    smart_call(

  File "/context/venv/lib/python3.12/site-packages/crunch/utils.py", line 233, in smart_call

    return function(**arguments)

           ^^^^^^^^^^^^^^^^^^^^^

  File "/context/code/main.py", line 720, in infer

    out_dir.mkdir(parents=True, exist_ok=True)

  File "/usr/local/lib/python3.12/pathlib.py", line 1311, in mkdir

    os.mkdir(self, mode)

PermissionError: [Errno 13] Permission denied: '/context/outputs'

command not exited correctly: 1

This program ran succesfully before

6

The cloud environment is very strict for security reasons.

You can only write in the directory provided via the prediction_directory_path argument to the infer() function.

You are also not permitted to write anywhere else, except in /tmp and the model (resources, provided via model_directory_path) directory.

I am guessing you did current_math / "../outputs" or similar?

If you want faster support, please contact us on the Discord as we are usually slowed to respond on the Forum.

Or at least create your own thread and not reuse another user’s thread with an unrelated issue. It make it harder for new users that have the same problem to search for it.

1 Like