The world's cleanest AutoML library ✨ - Do hyperparameter tuning with the right pipeline abstractions to write clean deep learning production pipelines. Let your pipeline steps have hyperparameter spaces. Design steps in your pipeline like components. Compatible with Scikit-Learn, TensorFlow, and most other libraries, frameworks and MLOps environments.
Describe the bug
Multithreading error when using processes instead of just threads.
To Reproduce
TypeError
cannot pickle 'weakref' object
File "/Users/guic/Documents/Neuraxle/neuraxle/distributed/streaming.py", line 197, in start
p.start()
File "/Users/guic/Documents/Neuraxle/neuraxle/distributed/streaming.py", line 519, in transform_data_container
step.start(context)
File "/Users/guic/Documents/Neuraxle/neuraxle/base.py", line 1287, in handle_transform
data_container = self.transform_data_container(data_container, context)
File "/Users/guic/Documents/Neuraxle/neuraxle/base.py", line 3537, in transform
data_container = self.handle_transform(data_container, context)
File "/Users/guic/Documents/Neuraxle/testing/test_streaming.py", line 323, in test_queued_pipeline_with_savers (Current frame)
outputs = p.transform(list(range(100)))
Expected behavior
No error
Suggested Fix
Investigate on Python 3.8 to 3.9 changes, and if it can be caused by Mac OS.
Additional context
Never tried on Mac OS Python 3.9 before. Coming from 3.8.
Describe the bug Multithreading error when using processes instead of just threads.
To Reproduce
Expected behavior No error
Suggested Fix Investigate on Python 3.8 to 3.9 changes, and if it can be caused by Mac OS.
Additional context Never tried on Mac OS Python 3.9 before. Coming from 3.8.