使用groupby对绘图数据进行分类

Question

我有下面的dataframe，我正在尝试在分组数据之后绘制值和误差线

import pandas as pd
from pprint import pprint


def classify_data():
    race = ['race1','race1','race1','race1','race2','race2','race2', 'race2']
    qualifier = ['last','first','first','first','last','last','first','first']
    participant = ['rat','rat','cat','cat','rat','dog','dog','dog']
    avgsteps =  ['223','321', '445', '165', '216', '127', '122', '214']
    measureerror = ['0.02', '0.01', '0.01', '0.05', '0.02', '0.01', '0.01', '0.05']
    id = ['a', ' b', 'c', 'd', 'e', 'f', 'g', 'h']
    df = pd.DataFrame(
        {'race': race,
         'qualifier': qualifier,
         'participant': participant,
         'avgsteps' : avgsteps,
         'measureerror': measureerror
        }
    )

我想将数据分组如下：

 race qualifier participant avgsteps measureerror  id

0  race1     first         rat      321         0.01   b
1  race1     first         cat      445         0.01   c
2  race1     first         cat      165         0.05   d
3  race2     first         dog      122         0.01   g
4  race2     first         dog      214         0.05   h

我打算为每个种族类型创建单独的绘图。 例如，race 1的绘图，其id位于x轴上，avgsteps和 y轴上measureerror的误差线。我想根据 participant类型

首先，我试着

   df2 = (df.groupby(['race', 'qualifier'])
           ) 

这将返回一个对象，我不知道如何继续

编辑：

  df = df[df.qualifier.str.contains('first')]

给予

1  race1     first         rat      321         0.01   b
2  race1     first         cat      445         0.01   c
3  race1     first         cat      165         0.05   d
6  race2     first         dog      122         0.01   g
7  race2     first         dog      214         0.05   h

但是，我不知道如何直接使用groupby

我想就如何进行征求意见

Answer 1

我没用groupby就找到了另一个选择

import pandas as pd
from pprint import pprint
from plotly.offline import download_plotlyjs, init_notebook_mode, plot, iplot
import plotly.graph_objs as go


def classify_data():
    race = ['race1','race1','race1','race1','race2','race2','race2', 'race2']
    qualifier = ['last','first','first','first','last','last','first','first']
    participant = ['rat','rat','cat','cat','rat','dog','dog','dog']
    avgsteps =  ['223','321', '445', '165', '216', '127', '122', '214']
    measureerror = ['20', '30', '121', '55', '21', '121', '100', '5']
    id = ['a', ' b', 'c', 'd', 'e', 'f', 'g', 'h']
    df = pd.DataFrame(
        {'race': race,
         'qualifier': qualifier,
         'participant': participant,
         'avgsteps' : avgsteps,
         'measureerror': measureerror,
         'id' : id
        }
    )
    df = df[df.qualifier.str.contains('first')]
    met_sids = df.race.unique()
    pprint(met_sids)
    colorsIdx = {'rat': 'rgb(215,48,39)', 'cat': 'rgb(215,148,39)', 'dog':'rgb(215,148,20)'}
    cols = df['participant'].map(colorsIdx)

    for met_sid in met_sids:
        temp_df = df[df.race.str.contains(f'{met_sid}')]
        pprint(temp_df)
        data = [
            go.Scatter(
                x=temp_df['id'],
                y=temp_df['avgsteps'],
                error_y=dict(
                    type='data',
                    array=temp_df['measureerror'],
                    visible=True,
                ),
                mode='markers',
                marker=dict(size=15, color=cols)

            )
        ]
        plot(data, filename='error-bar.html')

classify_data()

唯一的问题是误差线和数据点的颜色不同。有没有关于如何为值和误差条指定相同颜色的建议

我也很高兴知道其他的方法