1. 读取文件“linear_regression”，并将数据分为两部分：前2/3作为训练集，后2/3作为测试集2. 分别输出训练集中x列和y列的平均数、中位数和标准差；输出x列和y列的协方差；输出x列和y列的皮尔逊相关系数。3. 请打印出训练集中数据的散点图。

首先，为了完成这个任务，你需要使用Python的数据处理库，比如pandas用于读取文件并划分数据，numpy计算统计量，以及matplotlib绘制散点图。假设"linear_regression.csv"是一个包含两列数据（x和y）的CSV文件：

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

# 1. 读取文件，并划分数据
data = pd.read_csv('linear_regression.csv')
n_rows = len(data)
train_size = int(2 * n_rows / 3)
train_data = data[:train_size]
test_data = data[train_size:]

# 2. 计算训练集统计信息
mean_x_train = train_data['x'].mean()
median_x_train = train_data['x'].median()
std_dev_x_train = train_data['x'].std()

mean_y_train = train_data['y'].mean()
median_y_train = train_data['y'].median()
std_dev_y_train = train_data['y'].std()

cov_xy_train = train_data[['x', 'y']].cov().iloc[0, 1]  # 协方差
pearson_corr_train = train_data['x'].corr(train_data['y'])  # 相关系数

print(f"Training set x statistics:")
print(f"Mean: {mean_x_train}, Median: {median_x_train}, Standard Deviation: {std_dev_x_train}")
print(f"Training set y statistics:")
print(f"Mean: {mean_y_train}, Median: {median_y_train}, Standard Deviation: {std_dev_y_train}")
print(f"Covariance: {cov_xy_train}")
print(f"Pearson Correlation Coefficient: {pearson_corr_train}")

# 3. 绘制训练集散点图
plt.scatter(train_data['x'], train_data['y'])
plt.xlabel('X')
plt.ylabel('Y')
plt.title('Scatter plot of Training Data')
plt.show()