使用 Seaborn 或 Matplotlib 生成折线图，其中：年为色调，月为 X 轴，浮点列为 Y 轴

Question

The problem is I'm trying to generate a line graph using seaborn.lineplot() function, but I can't seem to find a way to generate a line graph like the one below:

https://i.stack.imgur.com/zUKog.png

My dataset has the following columns: Year, Month, Day, Units, Price per Unit, Sales. I've used the groupby function from pandas to sum the sales for each year in each month.

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

years_list = [date.strftime('%Y/%#-m/%#-d') for date in pd.date_range('01/01/2000', '31/12/2019')]

data = {
    'year': [int(date.split('/')[0]) for date in years_list],
    'month': [int(date.split('/')[1]) for date in years_list],
    'day': [int(date.split('/')[2]) for date in years_list],
    'units': [np.random.randint(1,25) for turns in range(len(years_list))],
    'price_per_unit': [np.random.uniform(10, 100) for turns in range(len(years_list))]
}

df = pd.DataFrame(data)
df['sales'] = df['price_per_unit'] * df['units']
each_month = df.groupby(['year', 'month'])['sales'].sum().reset_index()

Initially, I thought that using the code sns.lineplot(x='month', y='sales', hue='year', data=each_month) would automatically generate my desired graph, but it generated a confusing graph instead.

Anyone who has a solution for me? even if it isn't by using seaborn but matplotlib.

Answer 1

Yes, that line indeed creates what you want. However, for numeric hue, sns automatically colors your line according to a continuous colormap. To overcome this, you can convert your year to string type:

each_month['year'] = 'Y' + each_month['year'].astype(str)
plt.figure(figsize=(10,6))
sns.lineplot(x='month', y='sales', hue='year', data=each_month)

Output:

Answer 2

Set month as index and then groupby year , then just plot sales .

each_month.set_index('month').groupby('year').sales.plot(legend=True, figsize=(10, 6))