ValueError: Unable to configure handler 'distributed_logfile'

[ststevanovic]

Describe the bug

Trying to fit the model with simple dataset of shape (40, 4) with 3 Numerical 1 Categorical columns
Running on google colab (might be the issue...). Getting logging-related error.

To Reproduce

dataset.sample(6)
X = array([[165349.2, 136897.8, 471784.1, 'New York'],
[162597.7, 151377.59, 443898.53, 'California'],
[153441.51, 101145.55, 407934.54, 'Florida'],
[144372.41, 118671.85, 383199.62, 'New York'],
[142107.34, 91391.77, 366168.42, 'Florida'],
[131876.9, 99814.71, 362861.36, 'New York']])
y = array([ 96778.92, 96479.51, 105733.54, 96712.8 , 124266.9 , 155752.6])

model= AutoSklearnRegressor(
time_left_for_this_task=180,
per_run_time_limit=60,
ensemble_kwargs = {
'ensemble_size': 1
},
)
model.fit(X,y, feat_type = ['Numerical', 'Numerical', 'Numerical', 'Categorical'])

Expected behavior

Successful model fitting with automatic one-hot encoding

Actual behavior, stacktrace or logfile

---

FileNotFoundError Traceback (most recent call last)
/usr/lib/python3.10/logging/config.py in configure(self)
564 try:
--> 565 handler = self.configure_handler(handlers[name])
566 handler.name = name

12 frames
FileNotFoundError: [Errno 2] No such file or directory: '/tmp/auto-sklearn_tmp_4c47c997-e081-11ee-8586-0242ac1c000c/distributed.log'

The above exception was the direct cause of the following exception:

ValueError Traceback (most recent call last)
/usr/lib/python3.10/logging/config.py in configure(self)
570 deferred.append(name)
571 else:
--> 572 raise ValueError('Unable to configure handler '
573 '%r' % name) from e
574

ValueError: Unable to configure handler 'distributed_logfile'

same error in both dataframe and numpy array format

Environment and installation:

• Google colab
• Python 3.10.12
• Auto-sklearn version 0.15.0

[fuqiao14]

I also encountered the same problem. Have you resolved it?

[andreaschandra]

have you resolved it?