在 plotly express 中添加另一个数据框作为注释

Question

我使用 UMAP 可视化数据，但无法添加适当的注释。如何使用相同长度的另一个数据框将悬停文本添加到绘图表达散点图？ 据我了解，我只能从data_2d 那里指定一列。我可以从another_df 获取所有行作为注释吗？

import plotly.express as px
def scatter(data_2d, labels, another_df):
    c = dict(zip(labels.unique, px.colors.qualitative.G10)) # make the same colors for another data
    fig = px.scatter(
        data_2d, x=0, y=1,
        color=labels,
        color_discrete_map=c,
        text=another_df, # here I'm stuck
        # expected annotation 
      #  column1: 57575874
      #  column2: 0.4545
      #  columnN: ....

        # hover_data  awaits for labels from data_2d and it doesn't work
        # text is constant, I see just a mess of text
    )

    fig.update_traces(marker=dict(size=5, opacity=0.7))

    fig.show()

Answer 1

• 您的示例代码
  • 没有定义数据结构，来自 cmets 我已经暗示了代码中的内容
  • x 和 y 作为值在语法上不正确，已修复从 UMAP 提取到 2D numpy 数组
• 解决方案
  • 您已定义 another_df 与 data_2d 的长度相同。一种选择是在通过转换之前使用所有列的数据框
  • labels 参数确实是多余的，它是 another_df 的一部分
  • 您要定义 hover 输入，已定义 hover_name 和 hover_data。使用了 another_df 中的所有列

import plotly.express as px
import pandas as pd
import numpy as np
import umap.umap_ as umap
from sklearn.datasets import load_digits
from sklearn.model_selection import train_test_split
from sklearn.preprocessing import StandardScaler

def scatter(data_2d, labels, another_df):
    c = dict(zip(labels.unique(), px.colors.qualitative.G10)) # make the same colors for another data
    fig = px.scatter(
        another_df, 
        x=data_2d[:,0],y=data_2d[:,1],
        color=labels,
        color_discrete_map=c,
        hover_name="island",
        hover_data=another_df.columns
    )
    fig.update_traces(marker=dict(size=5, opacity=0.7))
    fig.show()

    
penguins = pd.read_csv("https://github.com/allisonhorst/palmerpenguins/raw/5b5891f01b52ae26ad8cb9755ec93672f49328a8/data/penguins_size.csv")
data = penguins.loc[:,["culmen_length_mm","culmen_depth_mm","flipper_length_mm","body_mass_g",]].dropna()
scaled_penguin_data = StandardScaler().fit_transform(data.values)

reducer = umap.UMAP()
embedding = reducer.fit_transform(scaled_penguin_data)
scatter(embedding, penguins.loc[data.index, "sex"].fillna("UNKNOWN"), penguins.loc[data.index])