Closed
Description
Describe the bug
When calling cross_val_score with a sparse data matrix X and a RandomForestClassifier with n_jobs=-1, there is a weird interaction with joblib and memmapping that makes the buffer from X read-only, breaking the cython code for the tree construction but it is weird as it only appears with the cross_validate function, and not when calling the classifier alone, while n_jobs=1 for the cross val function so joblib should not enter the play here...
Steps/Code to Reproduce
from scipy.sparse import csr_matrix
from sklearn.datasets import make_classification
from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import cross_val_score
X, y = make_classification(10000, n_features=200)
X = csr_matrix(X, copy=True)
clf = RandomForestClassifier(n_jobs=-1)
cross_val_score(clf, X, y)Expected Results
Working code
Actual Results
ValueError:
All the 5 fits failed.
It is very likely that your model is misconfigured.
You can try to debug the error by setting error_score='raise'.
Below are more details about the failures:
--------------------------------------------------------------------------------
5 fits failed with the following error:
joblib.externals.loky.process_executor._RemoteTraceback:
"""
Traceback (most recent call last):
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/joblib/externals/loky/process_executor.py", line 428, in _process_worker
r = call_item()
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/joblib/externals/loky/process_executor.py", line 275, in __call__
return self.fn(*self.args, **self.kwargs)
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/joblib/_parallel_backends.py", line 620, in __call__
return self.func(*args, **kwargs)
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/joblib/parallel.py", line 288, in __call__
return [func(*args, **kwargs)
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/joblib/parallel.py", line 288, in <listcomp>
return [func(*args, **kwargs)
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/sklearn/utils/fixes.py", line 117, in __call__
return self.function(*args, **kwargs)
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/sklearn/ensemble/_forest.py", line 185, in _parallel_build_trees
tree.fit(X, y, sample_weight=curr_sample_weight, check_input=False)
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/sklearn/tree/_classes.py", line 889, in fit
super().fit(
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/sklearn/tree/_classes.py", line 379, in fit
builder.build(self.tree_, X, y, sample_weight)
File "sklearn/tree/_tree.pyx", line 147, in sklearn.tree._tree.DepthFirstTreeBuilder.build
File "sklearn/tree/_tree.pyx", line 173, in sklearn.tree._tree.DepthFirstTreeBuilder.build
File "sklearn/tree/_splitter.pyx", line 789, in sklearn.tree._splitter.BaseSparseSplitter.init
File "stringsource", line 660, in View.MemoryView.memoryview_cwrapper
File "stringsource", line 350, in View.MemoryView.memoryview.__cinit__
ValueError: buffer source array is read-only
"""
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/sklearn/model_selection/_validation.py", line 686, in _fit_and_score
estimator.fit(X_train, y_train, **fit_params)
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/sklearn/ensemble/_forest.py", line 474, in fit
trees = Parallel(
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/joblib/parallel.py", line 1098, in __call__
self.retrieve()
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/joblib/parallel.py", line 975, in retrieve
self._output.extend(job.get(timeout=self.timeout))
File "/home/temp/.local/miniconda/lib/python3.10/site-packages/joblib/_parallel_backends.py", line 567, in wrap_future_result
return future.result(timeout=timeout)
File "/home/temp/.local/miniconda/lib/python3.10/concurrent/futures/_base.py", line 458, in result
return self.__get_result()
File "/home/temp/.local/miniconda/lib/python3.10/concurrent/futures/_base.py", line 403, in __get_result
raise self._exception
ValueError: buffer source array is read-only
Versions
System:
python: 3.10.6 | packaged by conda-forge | (main, Aug 22 2022, 20:36:39) [GCC 10.4.0]
executable: /home/temp/.local/miniconda/bin/python3.10
machine: Linux-5.14.0-1054-oem-x86_64-with-glibc2.31
Python dependencies:
sklearn: 1.2.0
pip: 22.3
setuptools: 65.5.0
numpy: 1.23.5
scipy: 1.9.3
Cython: 0.29.32
pandas: 1.5.2
matplotlib: 3.6.2
joblib: 1.2.0
threadpoolctl: 3.1.0
Built with OpenMP: True
threadpoolctl info:
user_api: blas
internal_api: mkl
prefix: libmkl_rt
filepath: /home/temp/.local/miniconda/lib/libmkl_rt.so.2
version: 2022.1-Product
threading_layer: intel
num_threads: 4
user_api: openmp
internal_api: openmp
prefix: libomp
filepath: /home/temp/.local/miniconda/lib/libomp.so
version: None
num_threads: 8