How to use the mlopt.sklearn_tune.SVCOpt function in mlopt

# we can set the range of the parameters tuned manually
# more info about the parameters to be tuned for each algorithm
# can be found in the MLOpt class

# kNN
tune_range = {'n_neighbors':[10,20,50,100]}
knn = KNNOpt(X_slice,y_slice,params_cv=params_cv,tune_range=tune_range,model_name='knn_porto_seguro',save_dir=save_dir)
knn.tune_params()
print('Best model parameters:')
print(knn.best_model)
knn.save_model()

# SVC
tune_range = {'C':[0.01,0.1,1,10,100],
              'gamma':[0.01,0.1,1]}
svc = SVCOpt(X_slice,y_slice,params_cv=params_cv,tune_range=tune_range,model_name='svc_porto_seguro',save_dir=save_dir)
svc.tune_params()
print('Best model parameters:')
print(svc.best_model)
svc.save_model()

# we can also set the parameters that are not tuned

# random forest
params = {'criterion':'gini'}
tune_range = {'n_estimators':[50,100,200,500],
              'max_features':[0.5,0.75],
              'min_samples_leaf':[0.001,0.01]}
rf = RandomForestClassifierOpt(X_slice,y_slice,params=params,params_cv=params_cv,tune_range=tune_range,
                               model_name='rf_porto_seguro',save_dir=save_dir)
rf.tune_params()
print('Best model parameters:')