Week 4 数据可视化

import matplotlib.pyplot as plt
import pandas as pd

diamonds = pd.read_csv('data/05-eda-viz/diamonds.csv')

diamonds['price'].plot.hist(bins=50, figsize=(8, 4))
plt.title('Price distribution')
plt.xlabel('price')
plt.ylabel('count')
plt.show()

import matplotlib.pyplot as plt
import pandas as pd

diamonds = pd.read_csv('data/05-eda-viz/diamonds.csv')

diamonds['price'].plot.box(figsize=(5, 4))
plt.title('Price boxplot')
plt.show()

import matplotlib.pyplot as plt
import pandas as pd

midwest = pd.read_csv('data/05-eda-viz/midwest.csv')

midwest.groupby('state')['poptotal'].sum().plot.bar(figsize=(8, 4))
plt.title('Total population by state')
plt.xlabel('state')
plt.ylabel('population')
plt.show()

import pandas as pd

diamonds = pd.read_csv('data/05-eda-viz/diamonds.csv')

print('shape:', diamonds.shape)
print('price 五数概括:')
print(diamonds['price'].describe().round(2).to_string())
print()
print('cut 计数:')
print(diamonds['cut'].value_counts().to_string())
print()
print('按 cut 分组的 price 均值与中位数:')
print(diamonds.groupby('cut')['price'].agg(['count', 'mean', 'median']).round(2).to_string())

import os
import matplotlib.pyplot as plt
import pandas as pd

diamonds = pd.read_csv('data/05-eda-viz/diamonds.csv')
out_dir = 'projects/demo/04-viz'
os.makedirs(out_dir, exist_ok=True)

# 直方图：标题带样本量
fig, ax = plt.subplots(figsize=(8, 4))
ax.hist(diamonds['price'], bins=50, color='#4c72b0', edgecolor='white')
ax.set_title(f"Diamond price distribution (n={len(diamonds)})")
ax.set_xlabel('price (USD)')
ax.set_ylabel('count')
fig.tight_layout()
fig.savefig(os.path.join(out_dir, 'diamonds_price_hist.png'), dpi=300)
plt.close(fig)

# 箱线图：按 cut 分类对比
fig, ax = plt.subplots(figsize=(8, 4))
diamonds.boxplot(column='price', by='cut', ax=ax, grid=False)
ax.set_title(f'Price by cut (n={len(diamonds)})')
ax.set_xlabel('cut')
ax.set_ylabel('price (USD)')
fig.suptitle('')
fig.tight_layout()
fig.savefig(os.path.join(out_dir, 'diamonds_price_by_cut.png'), dpi=300)
plt.close(fig)

print('图表已保存到:', out_dir)

import pandas as pd

midwest = pd.read_csv('data/05-eda-viz/midwest.csv')
state_pop = midwest.groupby('state')['poptotal'].sum().sort_values(ascending=False)

print('midwest shape:', midwest.shape)
print('各州总人口:')
print(state_pop.to_string())
print()
print('各州县数量:')
print(midwest['state'].value_counts().to_string())

import os
import matplotlib.pyplot as plt
import pandas as pd

midwest = pd.read_csv('data/05-eda-viz/midwest.csv')
state_pop = midwest.groupby('state')['poptotal'].sum().sort_values(ascending=False)
out_dir = 'projects/demo/04-viz'
os.makedirs(out_dir, exist_ok=True)

fig, ax = plt.subplots(figsize=(8, 4))
state_pop.plot.bar(ax=ax, color='#55a868')
ax.set_title(f'Total population by state (n_counties={len(midwest)})')
ax.set_xlabel('state')
ax.set_ylabel('population')
ax.ticklabel_format(style='plain', axis='y')
ax.set_xticklabels(ax.get_xticklabels(), rotation=45)
fig.tight_layout()
fig.savefig(os.path.join(out_dir, 'midwest_population.png'), dpi=300)
plt.close(fig)

import matplotlib.pyplot as plt
import pandas as pd

energy = pd.read_csv('data/05-eda-viz/energy_dataset.csv')
energy['time'] = pd.to_datetime(energy['time'], errors='coerce', utc=True)
energy['date'] = energy['time'].dt.date
daily = energy.groupby('date')['total load actual'].mean()

daily.plot.line(figsize=(8, 4))
plt.title('Average daily total load')
plt.xlabel('date')
plt.ylabel('total load actual')
plt.show()

import pandas as pd

energy = pd.read_csv('data/05-eda-viz/energy_dataset.csv')
energy['time'] = pd.to_datetime(energy['time'], errors='coerce', utc=True)
energy['total load actual'] = pd.to_numeric(energy['total load actual'], errors='coerce')
energy = energy.dropna(subset=['time', 'total load actual']).sort_values('time')
energy['date'] = energy['time'].dt.date
daily = energy.groupby('date')['total load actual'].mean()

print('解析后行数:', len(energy))
print('时间范围:', energy['time'].min(), '到', energy['time'].max())
print('日均负荷样本量:', len(daily))
print('最近 5 天:')
print(daily.tail().round(2).to_string())

import os
import matplotlib.pyplot as plt
import pandas as pd

energy = pd.read_csv('data/05-eda-viz/energy_dataset.csv')
energy['time'] = pd.to_datetime(energy['time'], errors='coerce', utc=True)
energy['total load actual'] = pd.to_numeric(energy['total load actual'], errors='coerce')
energy = energy.dropna(subset=['time', 'total load actual']).sort_values('time')
energy['date'] = energy['time'].dt.date
daily = energy.groupby('date')['total load actual'].mean()

out_dir = 'projects/demo/04-viz'
os.makedirs(out_dir, exist_ok=True)

fig, ax = plt.subplots(figsize=(8, 4))
daily.tail(120).plot.line(ax=ax, color='#c44e52')
ax.set_title('Average daily total load, last 120 days (n=120)')
ax.set_xlabel('date')
ax.set_ylabel('total load actual')
fig.tight_layout()
fig.savefig(os.path.join(out_dir, 'energy_load.png'), dpi=300)
plt.close(fig)

print('图表已保存到:', out_dir)

import matplotlib.pyplot as plt
import pandas as pd

life = pd.read_csv('data/04-cleaning/Life_Expectancy_Data.csv')
life.columns = [c.strip() for c in life.columns]

year_mean = life.groupby('Year')['Life expectancy'].mean()
year_mean.plot()
plt.show()

import matplotlib.pyplot as plt
import pandas as pd

life = pd.read_csv('data/04-cleaning/Life_Expectancy_Data.csv')
life.columns = [c.strip() for c in life.columns]

life.groupby(['Year', 'Status'])['Life expectancy'].mean().unstack().plot()
plt.show()

import matplotlib.pyplot as plt
import pandas as pd

life = pd.read_csv('data/04-cleaning/Life_Expectancy_Data.csv')
life.columns = [c.strip() for c in life.columns]
year_mean = life.groupby('Year')['Life expectancy'].mean()
status_year = life.groupby(['Year', 'Status'])['Life expectancy'].mean().unstack()

fig, axes = plt.subplots(1, 2, figsize=(10, 4))
year_mean.plot(ax=axes[0])
status_year.plot(ax=axes[1])
plt.show()

import pandas as pd

life = pd.read_csv('data/04-cleaning/Life_Expectancy_Data.csv')
life.columns = [c.strip() for c in life.columns]

year_mean = life.groupby('Year')['Life expectancy'].mean()

print('年份趋势样本量:', len(year_mean))
print(year_mean.round(2).tail().to_string())
print()

status_year = life.groupby(['Year', 'Status'])['Life expectancy'].mean().unstack()
print('Status x Year 各分组样本量:')
print(status_year.notna().sum().to_string())
print()
print(status_year.round(2).tail().to_string())

import os
import matplotlib.pyplot as plt
import pandas as pd

life = pd.read_csv('data/04-cleaning/Life_Expectancy_Data.csv')
life.columns = [c.strip() for c in life.columns]
year_mean = life.groupby('Year')['Life expectancy'].mean()
status_year = life.groupby(['Year', 'Status'])['Life expectancy'].mean().unstack()

out_dir = 'projects/demo/04-viz'
os.makedirs(out_dir, exist_ok=True)

fig, axes = plt.subplots(nrows=2, ncols=1, figsize=(8, 6), sharex=True)

year_mean.plot.line(ax=axes[0], color='#c44e52', linewidth=2)
axes[0].set_title(f'Global mean life expectancy by year (n_years={len(year_mean)})')
axes[0].set_ylabel('life expectancy')

status_year.plot.line(ax=axes[1])
axes[1].set_title('Life expectancy by development status')
axes[1].set_xlabel('year')
axes[1].set_ylabel('life expectancy')
axes[1].legend(title='Status')

fig.tight_layout()
fig.savefig(os.path.join(out_dir, 'life_expectancy_combined.png'), dpi=300)
plt.close(fig)

print('图表已保存到:', out_dir)