[MRG + 1] Fix ValueError in LabelEncoder when using inverse_transform on unseen labels

[ghost]

Reference Issue

Fix #9812

[amueller]

looks good to me.

[amueller]

This len is the fix, right?

[ghost]

Correct - that's all that was needed to produce a sensible error 😄

[amueller]

I find the organization of the tests a bit weird but not your fault. The test that it actually works if they are present is way at the top of the file.

[ghost]

Happy to reorganise tomorrow if you are able to give me some pointers - I'm not very familiar with the testing structure of sklearn as this is my first issue.

[amueller]

it's fine, I think.

[lesteve]

LGTM, merging, thanks a lot @newey01c!

[jnothman]

This is missing a whats_new entry. I'll pull it into my 0.19.1 branch and write an entry there

[vdaita]

The issue appears to be persistent - I am using LabelEncoder. Here is my stack trace:

 File "ann.py", line 71, in <module>
    X_train, X_test, y_train, y_test = get_dataset("Churn_Modelling.csv", 3, 13, 13)
  File "ann.py", line 28, in get_dataset
    encoder.fit(labels)
  File "/home/yolopc/.local/lib/python3.5/site-packages/sklearn/preprocessing/label.py", line 96, in fit
    self.classes_ = np.unique(y)
  File "/home/yolopc/.local/lib/python3.5/site-packages/numpy/lib/arraysetops.py", line 210, in unique
    return _unique1d(ar, return_index, return_inverse, return_counts)
  File "/home/yolopc/.local/lib/python3.5/site-packages/numpy/lib/arraysetops.py", line 277, in _unique1d
    ar.sort()
ValueError: The truth value of an array with more than one element is ambiguous. Use a.any() or a.all()

Do you have any suggestions?

[jnothman]

As noted in #10552, this was accidentally not included in the 0.19.1 release. Using the development version of scikit-learn will make it work.

[vdaita]

I tried to follow the instructions on the website, but I could not import any scikit-learn libraries. I also ran pip install -e, but I got the following error: IndexError: list index out of range. Do you have any pointers? I am using Python 2.7 on Ubuntu 16.04.

On Sat, Feb 3, 2018 at 6:41 PM Joel Nothman ***@***.***> wrote: As noted in #10552 <#10552>, this was accidentally not included in the 0.19.1 release. Using the development version of scikit-learn will make it work. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#9816 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AV4hUWvS5UzS3P1532AQtZOjXe8bnh6Sks5tRO68gaJpZM4PfUrv> .

-- Vijay

[jnothman]

not sure without a full traceback what index was out of range. you can try pip install https://g <https://hith> ithub.com/scikit-learn/scikit-learn/archive/master.zip to install the latest development version.

…

[vdaita]

Thank you - I found an alternate solution.

On Sun, Feb 4, 2018 at 5:00 PM Joel Nothman ***@***.***> wrote: not sure without a full traceback what index was out of range. you can try pip install https://g <https://hith> ithub.com/scikit-learn/scikit-learn/archive/master.zip to install the latest development version. > — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#9816 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AV4hUXCZMFcf1ARIl9XGIRyNj2IdsGZDks5tRihmgaJpZM4PfUrv> .

-- Vijay