Error while loading data using joblib and import errors of the dependent fiels for models

[Rupesh-Darimisetti]

The repository not unable to load the basic import files used in model from the tools path and even appending with

sys.path.append(".")

it resolves importing error of

fromtools.email_preprocess import preprocess

after which it throwing error with joblib loading files as

python naive_bayes/nb_author_id.py 
Traceback (most recent call last):
  File "D:\02_learning\udacity\machine_learning\ud120-projects\naive_bayes\nb_author_id.py", line 24, in <module>
    features_train, features_test, labels_train, labels_test = preprocess()
                                                               ^^^^^^^^^^^^
  File "D:\02_learning\udacity\machine_learning\ud120-projects\tools\email_preprocess.py", line 31, in preprocess
    word_data = joblib.load(open("tools/word_data.pkl","rb"))
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "D:\02_learning\udacity\machine_learning\ud120-projects\venv\Lib\site-packages\joblib\numpy_pickle.py", line 648, in load
    obj = _unpickle(fobj)
          ^^^^^^^^^^^^^^^
  File "D:\02_learning\udacity\machine_learning\ud120-projects\venv\Lib\site-packages\joblib\numpy_pickle.py", line 577, in _unpickle
    obj = unpickler.load()
          ^^^^^^^^^^^^^^^^
  File "C:\Program Files\WindowsApps\PythonSoftwareFoundation.Python.3.11_3.11.2544.0_x64__qbz5n2kfra8p0\Lib\pickle.py", line 1213, in load
    dispatch[key[0]](self)
  File "C:\Program Files\WindowsApps\PythonSoftwareFoundation.Python.3.11_3.11.2544.0_x64__qbz5n2kfra8p0\Lib\pickle.py", line 1337, in load_string
    raise UnpicklingError("the STRING opcode argument must be quoted")
_pickle.UnpicklingError: the STRING opcode argument must be quoted

[Rupesh-Darimisetti]

@jaycode @13rac1 @nmb10 @richardkalehoff please try to fix the codebase in order to run the models of machine learning.

[jaycode]

Hi @Rupesh-Darimisetti, the code was set up to use Python 3.6.3. It looks like you are using 3.11, which rendered the unpickle feature to fail. I suggest creating a new virtual environment to run it.

[Rupesh-Darimisetti]

Even though i use python 3.6.8 version and reinstall the venv it's throwing the following error @jaycode

 python naive_bayes/nb_author_id.py 
Traceback (most recent call last):
  File "naive_bayes/nb_author_id.py", line 24, in <module>  
    features_train, features_test, labels_train, labels_test
 = preprocess()
  File ".\tools\email_preprocess.py", line 34, in preprocess

    word_data = joblib.load(words_file_handler)
  File "D:\02_learning\udacity\machine_learning\ud120-projec
ts\venv\lib\site-packages\joblib\numpy_pickle.py", line 577,
 in load
    obj = _unpickle(fobj)
  File "D:\02_learning\udacity\machine_learning\ud120-projec
ts\venv\lib\site-packages\joblib\numpy_pickle.py", line 506,
 in _unpickle
    obj = unpickler.load()
  File "C:\Users\rupes\AppData\Local\Programs\Python\Python3
6-32\lib\pickle.py", line 1050, in load
    dispatch[key[0]](self)
  File "C:\Users\rupes\AppData\Local\Programs\Python\Python3
6-32\lib\pickle.py", line 1174, in load_string
    raise UnpicklingError("the STRING opcode argument must b
e quoted")
_pickle.UnpicklingError: the STRING opcode argument must be 
quoted
(venv) 

[Keroshi]

I was also facing the same issue. But after a few stack overflow threads, I found the issue to be the Line separator format for both word_data.pkl and email_author.pkl was the one that brought about the errors. Changing it to LF from CRLF or CR might do the trick.

[mirandaO]

I tried everything that was suggested here, and even I used ChatGPT to try to solve the problem, but after expending a couple of a hours, I gave up. Since I actually want to spend time in learning ML concepts, not spending time in solving configuration or setup issues, I just used a basic solution.

My solution to be able to run the nb_author_id.py script without any library error, was to:
1-use Python 2.7.18
2-go back to the last commit before the Python 3 migration and run:
git checkout 9ee2a2e
3-change requirements.txt line "scipy==0.18.1" to "scipy>=0.18.1" and save changes
4-run C:\Python27\python.exe -m pip install -r requirements.txt
5-run C:\Python27\python.exe nb_author_id.py