使用破折号的多线图表中的无序 y 轴

Question

我是破折号新手，正在努力创建多线图。我有一个数据并使用破折号绘制它并创建了一个多线图表。下面给出了可重现的代码-

import pandas as pd

import plotly.express as px
from dash import Dash, dcc, html, callback, Output, Input

df = pd.DataFrame({
    'foo': ['ab','ab','ab','ab', 'es','es','es','es', 'la','la','la','la', 'kl','kl','kl','kl', 'ys','ys','ys','ys'],
    'months': [1,3,6,12,1,3,6,12,1,3,6,12,1,3,6,12,1,3,6,12],
    'metric': [60.25,70.96,104.72,194.33,60.27,72.52,117.66,234.53,61.17,72.30,118.84,239.15,60.76,72.53,118.39,239.00,59.00,65.41,78.04,123.49]
})

app = Dash()

app.layout = (html.Div([
    html.H1(children='Metric assessment', style={'textAlign':'center'}),
    html.Div(["Enter multiplier for metric    ", dcc.Input(
            id="some_number",
            type="number",
            placeholder=1, min = 1, max = 20, style={"margin-left": "15px"}
        ),
    dcc.Graph(id='graph-content')
])
])
)

@callback(
    Output('graph-content', 'figure'),
    Input("some_number", "value")
)
def update_graph(some_number):
    dff = df.copy(deep = True)
    if not some_number:
        some_number = 1
    
    dff['Total metric'] = dff['metric']*some_number
    dff['Total metric'] = dff['Total metric'].map('{:,.2f}'.format)
    # dff = dff.sort_values(by = 'Total metric')
    fig = px.line(dff, x='months', y='Total metric', color = 'foo', markers=True, title = 'Metric vs months')
    return fig

if __name__ == '__main__':
    app.run(debug=True)

当我看到该图时，y 轴值是无序的，如下所示 -

即使在“总指标”列上对数据帧进行排序后，问题仍然存在。我该怎么做才能消除这个问题并获得正确的图表？

Answer 1

在回调中，“总指标”列被格式化（即作为字符串），因此 Plotly 将图中的相应数据处理为分类数据而不是数字数据。您可以使用

pandas.DataFrame.astype()

来解决这个问题：

dff['Total metric'] = dff['Total metric'].map('{:,.2f}'.format).astype(float)