具有 50 多个创建的虚拟变量的堆叠条形图（百分比）？

• 使用此技术：在 Pandas DataFrame 中计算行百分比？
• 然后它变得非常简单地使用plotly express

import pandas as pd
import numpy as np
import plotly.express as px

S = 4
df = pd.DataFrame(
    {
        **{"date": pd.date_range("1-jan-2021", freq="MS", periods=12)},
        **{
            f"dummy_var{i+1}": np.random.choice(
                list(range(S)),
                12,
                p=[0.9 if i == 0 else 0.1 / (S - 1) for i in range(S)],
            )
            for i in range(50)
        },
    }
)

px.bar(
    df.set_index("date").pipe(lambda d: d.div(d.sum(axis=1), axis=0)).reset_index(),
    x="date",
    y=[c for c in df.columns if c != "date"],
).update_layout(xaxis={"dtick": "M1"}, yaxis={"tickformat":".2%"})