数据处理中遇到的函数

1.读数据
test_data=pd.read_csv(‘D:/dataset/Titanic/test.csv’)

2.看数据
train_data.head()看前五行
train_data.info()看统计

3.看哪些列有null
train_data.isnull().sum()

4.填充缺失值
train_data[‘Age’].fillna(train_data[‘Age’].mean(), inplace=True)(这里是填充平均值)

5.改变值（比如’male’变成0)
train_data[‘Sex’].replace(‘male’,1,inplace=True)
train_data[‘Embarked’].replace([‘S’,‘C’,‘Q’],[0,1,2],inplace=True)

6.网格搜索

rfc_best=RandomForestClassifier()#随机森林
params = {'n_estimators':list(range(50,300,50)),'max_depth':list(range(2,10,2))}
gsearch = GridSearchCV(rfc_best,params,n_jobs=-1,cv = 5,verbose=1)
gsearch.fit(train_data,train_target)

7.写答案

predictions =gsearch.predict(x_test)
submission = pd.DataFrame({'PassengerID' : test_data.PassengerId , 'Survived': predictions.astype(np.int32)})
submission.to_csv(r'D:/dataset/Titanic/rfc_best_submission.csv',index = False)

8.train_test_split

X_train,X_test, y_train, y_test =sklearn.model_selection.train_test_split(train_data,train_target,test_size=0.4, random_state=0,stratify=y_train)