Hyperparameters tuning with cross validation

Hello, I applied gridSearch (with cross validation) method to find the best parameters; then I fit again the model with the best parametes:

svc = svm.SVC(class_weight='balanced')
cv = StratifiedShuffleSplit(n_splits=5, test_size=0.3, random_state=109)
clf = GridSearchCV(svc, par, cv=cv)
search = clf.fit(Xtrain, ytrain)
b_par = search.best_params_
svc_best = svm.SVC(C=best_C, kernel=best_kernel, gamma=best_gamma, probability=True, 
svc_best.fit(X_train, y_train)
y_pred = svc_best.predict(X_test)

My question is: is there a way to re-fit the model with the addition of the cross validation?

I don’t understand the question, since you are already using the training set (which was previously used to run the cross-validation).
This question in unrelated to PyTorch, so you might get a faster and better answer in a scikit-learn discussion board.