[scikit-learn] Linear Discriminant Analysis with Cross Validation in Python

[Sebastian Raschka]

Hi, Loukas and Mahesh,
for LOOCV, you could e.g., use the LeaveOneOut class

```
from sklearn.discriminant_analysis import LinearDiscriminantAnalysis
from sklearn.model_selection import LeaveOneOut

loo = LeaveOneOut()
lda = LinearDiscriminantAnalysis()

test_fold_predictions = []

for train_index, test_index in loo.split(X):
    X_train, X_test = X[train_index], X[test_index]
    y_train, y_test = y[train_index], y[test_index]
    lda.fit(X_train, y_train)
    test_fold_predictions.append(lda.predict(X_test))
```

or you could pass the loo to the cross_val_score function directly:

```
from sklearn.model_selection import cross_val_score
cross_val_score(estimator=lda, X=X, y=y, cv=loo)
```


Best,
Sebastian


> On Mar 7, 2017, at 10:01 AM, Serafeim Loukas <seralouk at gmail.com> wrote:
> 
> Dear Mahesh,
> 
> Thank you for your response.
> 
> I read the documentation however I did not find anything related to cross-validation (leave one out).
> Can you give me a hint?
> 
> Thank you,
> S
> 
> .............................................
> Loukas Serafeim  
> University of Geneva
> email: seralouk at gmail.com
> 
> 
> 2017-03-07 10:56 GMT+01:00 Mahesh Kulkarni <maheshak04 at gmail.com>:
> Yes. Please see following link:
> 
> http://scikit-learn.org/stable/modules/generated/sklearn.discriminant_analysis.LinearDiscriminantAnalysis.html
> 
> On Tue, Mar 7, 2017 at 3:18 PM, Serafeim Loukas <seralouk at gmail.com> wrote:
> Dear scikit members,
> 
> 
> I would like to ask if there is any function that implements Linear Discriminant Analysis with Cross Validation (leave one out).
> 
> Thank you in advance,
> S
> 
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
> 
> 
> 
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn
> 
> 
> _______________________________________________
> scikit-learn mailing list
> scikit-learn at python.org
> https://mail.python.org/mailman/listinfo/scikit-learn