KeyError - Possible joblib version mismatch

[viniciusarruda]

Hi,

I'm trying to use your trained models, however, I am getting an error for the unigram model:

Traceback (most recent call last):
File "eval.py", line 216, in
main()
File "eval.py", line 206, in main
nltk_preds = predict_nltk(macmorpho['corpus'])
File "eval.py", line 101, in predict_nltk
tagger = joblib.load(folder + pkl_file)
File "/usr/local/lib/python3.8/dist-packages/joblib/numpy_pickle.py", line 587, in load
obj = _unpickle(fobj, filename, mmap_mode)
File "/usr/local/lib/python3.8/dist-packages/joblib/numpy_pickle.py", line 506, in _unpickle
obj = unpickler.load()
File "/usr/lib/python3.8/pickle.py", line 1212, in load
dispatchkey[0]
KeyError: 35

I found this answer at stack overflow: https://stackoverflow.com/questions/48948209/keyerror-when-loading-pickled-scikit-learn-model-using-joblib and maybe this can be the issue.

I was using joblib version 1.1.0 when I got the error. Downgrading to version 1.0.0 also throws the same error.

I think that the ideal scenario is to use the same version used to build and dump the model.