使用 pandas 数据框生成一系列图 [英] generate series of plots with pandas dataframe

问题描述

我必须生成一系列散点图(总共大约 100 个).

I have to generate a series of scatter plots (roughly 100 in total).

我创建了一个示例来说明问题.

I have created an example to illustrate the problem.

首先进行导入.

import pandas as pd

创建一个熊猫数据框.

 # Create dataframe
data = {'name': ['Jason', 'Jason', 'Tina', 'Tina', 'Tina', 'Jason', 'Tina'],
        'report_value': [4, 24, 31, 2, 3, 5, 10],
        'coverage_id': ['m1', 'm2', 'm3', 'm4', 'm5', 'm6', 'm7']}
df = pd.DataFrame(data)
print(df)

输出:

  coverage_id   name  report_value
0          m1  Jason             4
1          m2  Jason            24
2          m3   Tina            31
3          m4   Tina             2
4          m5   Tina             3
5          m6  Jason             5
6          m7   Tina            10

目标是在不使用 for 循环的情况下生成两个散点图.标题中应显示此人的姓名，Jason 或 Tina.report_value 应该在两个图中的 y 轴上，coverage_id(它是一个字符串)在 x 轴上.

The goal is generate two scatter plots without using a for-loop. The name of the person, Jason or Tina, should be displayed in the title. The report_value should be on the y-axis in both plots and the coverage_id (which is a string) on the x-axis.

我想我应该从:

df.groupby('name')

然后我需要将操作应用到每个组.

Then I need to apply the operation to every group.

这样我就有了按名称分组的数据框.我不知道如何继续并让 Python 为我制作这两个图.

This way I have the dataframe grouped by their names. I don't know how to proceed and get Python to make the two plots for me.

非常感谢您的帮助.

推荐答案

我认为你可以使用这个解决方案，但是首先必须将 string 列转换为数字，plot 和最后一组 xlabels:

I think you can use this solution, but first is necessary convert string column to numeric, plot and last set xlabels:

import matplotlib.pyplot as plt

u, i = np.unique(df.coverage_id, return_inverse=True)
df.coverage_id = i

groups = df.groupby('name')

# Plot
fig, ax = plt.subplots()
ax.margins(0.05) # Optional, just adds 5% padding to the autoscaling
for name, group in groups:
    ax.plot(group.coverage_id, 
            group.report_value, 
            marker='o', 
            linestyle='', 
            ms=12, 
            label=name)

ax.set(xticks=range(len(i)), xticklabels=u)
ax.legend()

plt.show()

另一个 seaborn 解决方案，带有 seaborn.pairplot:

Another seaborn solution with seaborn.pairplot:

import seaborn as sns

u, i = np.unique(df.coverage_id, return_inverse=True)
df.coverage_id = i

g=sns.pairplot(x_vars=["coverage_id"], y_vars=["report_value"], data=df, hue="name", size=5)
g.set(xticklabels=u, xlim=(0, None))

这篇关于使用 pandas 数据框生成一系列图的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！