Closed as not planned
Description
Description
Loading a dataset that has been downloaded to ~/scikit_learn_data some time in the past can throw unpickling errors after updating scikit-learn. Possible solutions might be:
- clearing the dataset cache after updating
- including the scikit-learn version used to serialize the dataset objects
- serializing with something besides
pickle
Deleting the data folder resolves this issue.
Steps/Code to Reproduce
python examples/impute/plot_iterative_imputer_variants_comparison.py
Expected Results
No error is thrown.
Actual Results
An unpickling error is thrown when trying to find the _joblib module in sklearn.externals.
Traceback (most recent call last):
File "examples/impute/plot_iterative_imputer_variants_comparison.py", line 61, in <module>
X_full, y_full = fetch_california_housing(return_X_y=True)
File "/Users/dillon/githubpackages/scikit-learn/sklearn/datasets/california_housing.py", line 133, in fetch_california_housing
cal_housing = _refresh_cache([filepath], 6)
File "/Users/dillon/githubpackages/scikit-learn/sklearn/datasets/base.py", line 930, in _refresh_cache
data = tuple([joblib.load(f) for f in files])
File "/Users/dillon/githubpackages/scikit-learn/sklearn/datasets/base.py", line 930, in <listcomp>
data = tuple([joblib.load(f) for f in files])
File "/Users/dillon/anaconda/envs/sklearn-dev/lib/python3.7/site-packages/joblib/numpy_pickle.py", line 598, in load
obj = _unpickle(fobj, filename, mmap_mode)
File "/Users/dillon/anaconda/envs/sklearn-dev/lib/python3.7/site-packages/joblib/numpy_pickle.py", line 526, in _unpickle
obj = unpickler.load()
File "/Users/dillon/anaconda/envs/sklearn-dev/lib/python3.7/pickle.py", line 1085, in load
dispatch[key[0]](self)
File "/Users/dillon/anaconda/envs/sklearn-dev/lib/python3.7/pickle.py", line 1373, in load_global
klass = self.find_class(module, name)
File "/Users/dillon/anaconda/envs/sklearn-dev/lib/python3.7/pickle.py", line 1423, in find_class
__import__(module, level=0)
ModuleNotFoundError: No module named 'sklearn.externals._joblib'
Versions
System:
python: 3.7.3 (default, Mar 27 2019, 16:54:48) [Clang 4.0.1 (tags/RELEASE_401/final)]
executable: /Users/dillon/anaconda/envs/sklearn-dev/bin/python
machine: Darwin-17.7.0-x86_64-i386-64bit
BLAS:
macros: SCIPY_MKL_H=None, HAVE_CBLAS=None
lib_dirs: /Users/dillon/anaconda/envs/sklearn-dev/lib
cblas_libs: mkl_rt, pthread
Python deps:
pip: 19.1.1
setuptools: 41.0.1
sklearn: 0.22.dev0
numpy: 1.16.4
scipy: 1.2.1
Cython: 0.29.11
pandas: 0.24.2
matplotlib: 3.1.0
joblib: 0.13.2