2.1.3

# Load the data

file_path = 'finance数据集.csv'
data = pd.__________(file_path)

---

data = pd.read_csv(file_path)

# 识别数值列用于箱线图

numeric_cols = data.select_dtypes(include=['float64', 'int64']).__________

---

numeric_cols = data.select_dtypes(include=['float64', 'int64']).columns

代码解释:

select_dtypes筛选数据类型

箱线图，查看数据分布是否有异常值

数值列，columns（键盘打出col后点击“ Tab”按键）

# 使用IQR处理异常值

Q1 = data[numeric_cols].__________(0.25)

Q3 = data[numeric_cols].quantile(0.75)

---

Q1 = data[numeric_cols].quantile(0.25)

代码解释：类似于一个箱子，

# 检查重复值

duplicates = data_cleaned.duplicated()

num_duplicates = duplicates.sum()

scaler = MinMaxScaler()

data_cleaned[numeric_cols] = scaler.__________(data_cleaned[numeric_cols])

---

data_cleaned[numeric_cols] = scaler.fit_transform(data_cleaned[numeric_cols])

代码解释:

# 将SeriousDlqin2yrs设为目标变量
target_variable = '__________'

---

target_variable = 'SeriousDlqin2yrs'

# 定义特征和目标

X = data_cleaned.drop(columns=[__________])
y = data_cleaned[__________]

---

X = data_cleaned.drop(columns=[target_variable])
y = data_cleaned[target_variable]

代码解释:target_variable 通常指的是目标变量（也称为因变量、标签或响应变量）

# 划分数据

X_train, X_test, y_train, y_test = __________(X, y, test_size=0.2, random_state=42)

---
X_train, X_test, y_train, y_test =train_test_split(X, y, test_size=0.2, random_state=42)

代码解释:划分数据集train_test_split

# 保存清洗后的数据到CSV

cleaned_file_path = '2.1.3_cleaned_data.csv'
data_cleaned.to_csv(cleaned_file_path, index=False)