Python数据分析：洞察数据背后的模式

import pandas as pd# 加载 CSV 文件df = pd.read_csv("data.csv")# 清洗和准备数据df = df.dropna()# 删除缺失值df["column"] = df["column"].astype("cateGory")# 转换数据类型# 合并数据集df2 = pd.read_csv("data2.csv")df = pd.merge(df, df2, on="id")

import numpy as np# 创建一个 NumPy 数组arr = np.array([1, 2, 3, 4, 5])# 数组操作arr_mean = np.mean(arr)# 计算平均值arr_sum = np.sum(arr)# 计算总和

import matplotlib.pyplot as plt# 创建一个散点图plt.scatter(df["x"], df["y"])plt.xlabel("x")plt.ylabel("y")plt.show()

import seaborn as sns# 创建一个热力图sns.heatmap(df.corr())# 计算相关矩阵并绘制热力图plt.show()

from sklearn.model_selection import train_test_splitfrom sklearn.linear_model import LinearRegression# 划分训练和测试集X_train, X_test, y_train, y_test = train_test_split(df[["x", "y"]], df["z"])# 训练线性回归模型model = LinearRegression()model.fit(X_train, y_train)# 评估模型score = model.score(X_test, y_test)# 计算准确率

import Tensorflow as tf# 创建一个神经网络模型model = tf.keras.Sequential([tf.keras.layers.Dense(128, activation="relu"),tf.keras.layers.Dense(1, activation="sigmoid")])# 训练模型model.compile(optimizer="adam", loss="binary_crossentropy", metrics=["accuracy"])model.fit(X_train, y_train, epochs=10)# 评估模型loss, accuracy = model.evaluate(X_test, y_test)