How to implement gridsearchCV for onevsrestclassifier of LogisticRegression classifier?

Question

parameters = [{'C': [10**-2, 10**-1, 10**0,10**1, 10**2, 10**3]}]

model_tunning = GridSearchCV(OneVsRestClassifier(LogisticRegression(penalty='l1')), param_grid=parameters,scoring="f1")
model_tunning.fit(x_train_multilabel, y_train)




ValueError                                Traceback (most recent call last)
<ipython-input-38-5d5850fe8978> in <module>()
  2 
  3 model_tunning = GridSearchCV(OneVsRestClassifier(LogisticRegression(penalty='l1')), param_grid=parameters,scoring="f1")
----> 4 model_tunning.fit(x_train_multilabel, y_train)

ValueError: Invalid parameter C for estimator OneVsRestClassifier(estimator=LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,
      intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,
      penalty='l1', random_state=None, solver='liblinear', tol=0.0001,
      verbose=0, warm_start=False),
      n_jobs=1). Check the list of available parameters with `estimator.get_params().keys()

Answer 1

When you use nested estimators with grid search you can scope the parameters with __ as a separator. In this case the LogisticRegression model is stored as an attribute named estimator inside the OneVsRestClassifier model:

from sklearn.model_selection import GridSearchCV
from sklearn.linear_model import LogisticRegression
from sklearn.multiclass import OneVsRestClassifier
tuned_parameters = [{'estimator__C': [100, 10, 1, 0.1, 0.01, 0.001, 0.0001]}]
Find Optimal C by grid search
log_reg_clf = OneVsRestClassifier(LogisticRegression())
logistic_gs = GridSearchCV(log_reg_clf, tuned_parameters,scoring = 'f1_micro', cv=3)
logistic_gs.fit(x_train_bow, y_train)
print(logistic_gs.best_estimator_)

Answer 2

You can see I have set up a basic pipeline here using GridSearchCV, tf-idf, Logistic Regression and OneVsRestClassifier. In the param_grid, you can set 'clf__estimator__C' instead of just 'C'

import numpy as np
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.multiclass import OneVsRestClassifier
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import GridSearchCV
from sklearn.pipeline import Pipeline
tfidf_vectorizer = TfidfVectorizer(smooth_idf=True)
log_reg_clf = OneVsRestClassifier(
  estimator=LogisticRegression(
    intercept_scaling=1,
    class_weight='balanced',
    random_state=0
  )
)
Create regularization hyperparameter space
C = np.logspace(0, 4, 10)
param_grid = [{
  'vect__ngram_range': [(1, 1), (1, 2), (1, 3), (1, 4)],
  'vect__max_features': (None, 5000, 10000, 50000),
  'vect__norm': ['l1','l2'],
  'clf__estimator__C': C,
  'clf__estimator__penalty': ['l1','l2']
}]
log_reg_clf_tfidf = Pipeline([
  ('vect', tfidf_vectorizer),
  ('clf', log_reg_clf)
])
print(log_reg_clf_tfidf.get_params().keys())
gs_logReg_tfidf = GridSearchCV(
  log_reg_clf_tfidf,
  param_grid,
  scoring='accuracy',
  cv=5,
  verbose=1,
  n_jobs=-1
)
gs_logReg_tfidf.fit(X_train, y_train)
print("The best parameters: \n", gs_logReg_tfidf.best_params_)
print("The best score: \n", gs_logReg_tfidf.best_score_)
df_test_predicted_idf = gs_logReg_tfidf.predict(X_test)