pip install matplotlib pandas requests beautifulsoup4

import requests
from bs4 import BeautifulSoup
import pandas as pd
 
def crawl_job_data():
    headers = {'User-Agent': 'Mozilla/5.0'}
    jobs = []
    
    for page in range(1, 6):  # 爬取前5頁
        url = f"https://example.com/jobs?page={page}&keyword=Python"
        response = requests.get(url, headers=headers)
        soup = BeautifulSoup(response.text, 'html.parser')
        
        # 假設(shè)每個(gè)職位在class為'job-item'的div中
        for item in soup.find_all('div', class_='job-item'):
            title = item.find('h2').text.strip()
            salary = item.find('span', class_='salary').text.strip()
            city = item.find('span', class_='location').text.strip()
            jobs.append({'title': title, 'salary': salary, 'city': city})
    
    df = pd.DataFrame(jobs)
    df.to_csv('job_data.csv', index=False)
    return df
 
data = crawl_job_data()

def clean_salary(salary_str):
    try:
        # 提取數(shù)字部分，例如"15-20K/月"→[15,20]
        parts = salary_str.replace('K', '').replace('/月', '').split('-')
        if len(parts) == 2:
            return (float(parts[0]) + float(parts[1])) / 2  # 取平均值
        return float(parts[0])
    except:
        return None
 
data['salary'] = data['salary'].apply(clean_salary)
data = data.dropna(subset=['salary'])  # 刪除薪資為空的記錄

city_counts = data['city'].value_counts().head(10)  # 取前10個(gè)城市

import matplotlib.pyplot as plt
 
plt.figure(figsize=(12, 6))
city_counts.plot(kind='bar', color='#4C72B0')
 
plt.title('Python崗位城市分布', fontsize=16)
plt.xlabel('城市', fontsize=12)
plt.ylabel('職位數(shù)量', fontsize=12)
plt.xticks(rotation=45)  # 旋轉(zhuǎn)x軸標(biāo)簽
plt.grid(axis='y', linestyle='--', alpha=0.7)
 
# 在柱子上方顯示數(shù)值
for i, v in enumerate(city_counts):
    plt.text(i, v + 50, str(v), ha='center')
 
plt.tight_layout()  # 自動(dòng)調(diào)整布局
plt.savefig('city_distribution.png', dpi=300)
plt.show()

# 假設(shè)數(shù)據(jù)中有'experience'字段
exp_salary = data.groupby('experience')['salary'].mean()
 
plt.figure(figsize=(10, 5))
exp_salary.plot(kind='line', marker='o', color='#DD8452')
 
plt.title('不同經(jīng)驗(yàn)要求的平均薪資', fontsize=14)
plt.xlabel('經(jīng)驗(yàn)要求', fontsize=11)
plt.ylabel('平均薪資(K)', fontsize=11)
plt.grid(True, linestyle=':', alpha=0.6)
 
# 標(biāo)記數(shù)據(jù)點(diǎn)
for x, y in zip(exp_salary.index, exp_salary):
    plt.text(x, y, f'{y:.1f}K', ha='center', va='bottom')
 
plt.show()

# 假設(shè)我們隨機(jī)生成一些經(jīng)驗(yàn)?zāi)晗迶?shù)據(jù)
import numpy as np
np.random.seed(42)
data['experience_years'] = np.random.randint(1, 11, size=len(data))
 
plt.figure(figsize=(10, 8))
plt.scatter(data['experience_years'], data['salary'], 
            alpha=0.6, color='#55A868', s=50)  # s控制點(diǎn)大小
 
plt.title('工作經(jīng)驗(yàn)與薪資關(guān)系', fontsize=15)
plt.xlabel('工作經(jīng)驗(yàn)(年)', fontsize=12)
plt.ylabel('薪資(K)', fontsize=12)
 
# 添加趨勢線
z = np.polyfit(data['experience_years'], data['salary'], 1)
p = np.poly1d(z)
plt.plot(data['experience_years'], p(data['experience_years']), 
         "r--", linewidth=1)
 
plt.grid(True)
plt.show()

# 創(chuàng)建薪資區(qū)間
bins = [0, 10, 15, 20, 30]
labels = ['10K以下', '10-15K', '15-20K', '20K以上']
data['salary_range'] = pd.cut(data['salary'], bins=bins, labels=labels)
 
range_counts = data['salary_range'].value_counts()
 
plt.figure(figsize=(8, 8))
plt.pie(range_counts, labels=range_counts.index, 
        autopct='%1.1f%%', startangle=90,
        colors=['#C44E52', '#CCB974', '#64776D', '#798E87'])
 
plt.title('Python崗位薪資分布', fontsize=15)
plt.tight_layout()
plt.show()

fig, axes = plt.subplots(2, 2, figsize=(14, 10))
fig.suptitle('Python崗位數(shù)據(jù)分析儀表盤', fontsize=18)
 
# 子圖1：城市分布
axes[0,0].bar(city_counts.index, city_counts.values, color='#4C72B0')
axes[0,0].set_title('城市分布')
axes[0,0].tick_params(axis='x', rotation=45)
 
# 子圖2：經(jīng)驗(yàn)-薪資散點(diǎn)
axes[0,1].scatter(data['experience_years'], data['salary'], 
                  alpha=0.5, color='#55A868')
axes[0,1].set_title('經(jīng)驗(yàn)與薪資關(guān)系')
 
# 子圖3：薪資范圍餅圖
axes[1,0].pie(range_counts, labels=None, autopct='%1.1f%%',
              colors=['#C44E52', '#CCB974', '#64776D', '#798E87'])
axes[1,0].set_title('薪資分布')
 
# 子圖4：經(jīng)驗(yàn)要求折線圖
exp_salary.plot(kind='line', marker='o', ax=axes[1,1], color='#DD8452')
axes[1,1].set_title('經(jīng)驗(yàn)要求與平均薪資')
 
plt.tight_layout()
plt.subplots_adjust(top=0.92)  # 調(diào)整標(biāo)題位置
plt.savefig('dashboard.png', dpi=300)
plt.show()

plt.style.use('seaborn')  # 使用seaborn樣式
# 或自定義樣式
plt.rcParams.update({
    'font.family': 'Microsoft YaHei',  # 中文支持
    'axes.titlesize': 14,
    'axes.labelsize': 12,
    'xtick.labelsize': 10,
    'ytick.labelsize': 10,
    'figure.figsize': (10, 6),
    'savefig.dpi': 300
})

# 模擬數(shù)據(jù)
phone_data = {
    'brand': ['Apple']*5 + ['Samsung']*5 + ['Huawei']*5,
    'model': ['iPhone13']*3 + ['iPhone13 Pro']*2 + 
              ['S22']*3 + ['S22 Ultra']*2 +
              ['P50']*3 + ['Mate50']*2,
    'price': [5999, 6799, 5499, 8999, 9799,
              4999, 5499, 6299, 7999, 8999,
              4499, 4999, 5299, 6799, 7499],
    'rating': [4.8, 4.9, 4.7, 4.9, 4.8,
               4.6, 4.7, 4.8, 4.9, 4.8,
               4.5, 4.6, 4.7, 4.8, 4.7]
}
 
df = pd.DataFrame(phone_data)

plt.figure(figsize=(10, 6))
df.boxplot(column='price', by='brand', 
           patch_artist=True,
           boxprops=dict(facecolor='#4C72B0', color='black'),
           medianprops=dict(color='white'))
 
plt.title('各品牌手機(jī)價(jià)格分布')
plt.suptitle('')  # 去除自動(dòng)生成的標(biāo)題
plt.xlabel('品牌')
plt.ylabel('價(jià)格(元)')
plt.grid(True, linestyle='--', alpha=0.6)
plt.show()

plt.figure(figsize=(10, 6))
brands = df['brand'].unique()
colors = {'Apple': '#4C72B0', 'Samsung': '#55A868', 'Huawei': '#C44E52'}
 
for brand in brands:
    subset = df[df['brand'] == brand]
    plt.scatter(subset['price'], subset['rating'], 
                s=100, label=brand, color=colors[brand])
 
plt.title('手機(jī)價(jià)格與評(píng)分關(guān)系')
plt.xlabel('價(jià)格(元)')
plt.ylabel('評(píng)分')
plt.legend()
plt.grid(True)
plt.show()

proxies = {
    'http': 'http://your-proxy-ip:port',
    'https': 'https://your-proxy-ip:port'
}
response = requests.get(url, headers=headers, proxies=proxies)

plt.rcParams['font.sans-serif'] = ['SimHei']  # Windows系統(tǒng)
plt.rcParams['axes.unicode_minus'] = False  # 解決負(fù)號(hào)顯示問題
# 或使用Mac/Linux的字體
# plt.rcParams['font.sans-serif'] = ['Arial Unicode MS']

plt.savefig('transparent.png', dpi=300, transparent=True)