Iris_classifier_logistic_svc

1. Claasification Model

 

predict discrete problems

1.1 Logistic Regression

• one of the classification models

• comes from LinearREgression, and nonlinearize y = wx+b to below function

• y = 1 / 1 + e^-z , z = wx+b, y∈（0, 1）

  

  

• if predict y > 0.5 => 1, else y => 0

• params in logistic regression is the same as linear regression, w = coef_, b = intercept_  
  

1.2 SVM / Support Vector Machine

• Zhihu: reference

 

2. Overfitting

3. Regulation

 

Code

iris_data = pd.read_csv(DATA_FILE)
X = iris_data[FEATURE_COL].values
y = iris_data['Species'].map(SPECIES_LABEL_DICT).values

X_train,X_test,y_train,y_test = train_test_split(X,y,test_size=1/3,random_state=10)

model_dict = {
    'kNN':KNeighborsClassifier(n_neighbors=10),
    # 'linear_reg':LinearRegression(), # linear_reg is not a classification model, R2 the bigger the better
    'logistic_reg':LogisticRegression(C=1e3),
    'SVC':SVC(C=1e3)
}

for model_name, model in model_dict.items():
    model.fit(X_train,y_train)
    accuracy = model.score(X_test,y_test)  # All the classification models give the accuracy
    print('Model:{} - Accuracy:{}'.format(model_name,accuracy))

# Results:
# Model:kNN - Accuracy:1.0
# Model:linear_reg - R2:0.9063358327734319 
# Model:logistic_reg - Accuracy:0.98
# Model:SVC - Accuracy:0.92

---

4. Cross Validation