Plotly override legend names without traces-CodePudding

I'm trying to change the legend names and I've found many answers but none of them works if you are not using traces. I learned to change the legend title but not the names. Here it is my code, hopefully somebody can help:

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

def camasRegion():  
    df = pd.read_csv('https://raw.githubusercontent.com/MinCiencia/Datos-COVID19/master/output/producto52/Camas_UCI.csv')
    dates = df.columns #for some reason I can't use this as x axis
    habilitadas = []
    ocupadasCovid = []
    ocupadasNoCovid = []
    region = 'Ñuble'

    for i in range(len(df)):
        if df.values[i][0] == region:
            if df.values[i][1] == 'Camas UCI habilitadas':          
                for x in range(len(df.columns)):
                    habilitadas.append(df.values[i][x])

            if df.values[i][1] == 'Camas UCI ocupadas COVID-19':            
                for x in range(len(df.columns)):
                    ocupadasCovid.append(df.values[i][x])

            if df.values[i][1] == 'Camas UCI ocupadas no COVID-19':         
                for x in range(len(df.columns)):
                    ocupadasNoCovid.append(df.values[i][x])     

    fig = px.line(df, 
        y = [habilitadas, ocupadasCovid, ocupadasNoCovid],
        x = range(len(fechas)), #I'm doing this because x = dates doesn't work
        title='Camas por región: (' region ')',)
        

    fig.add_annotation(
        x = 1, y = -0.1, 
        text = 'Fuente: Datos obtenidos desde el Ministerio de Ciencia.', 
        showarrow = False, xref='paper', yref='paper', 
        xanchor='right', yanchor='auto', xshift=0, yshift=-20
        )
    fig.update_layout(
        legend_title_text='Estado de cama',
        xaxis_title="Días",
        yaxis_title="Cantidad de camas",
        )
    fig.show()


camasRegion()

CodePudding user response：

You can't use dates, because df.columns besides dates contains strings Region and Serie. I simplified your code, it also made it more readable. Try this and let me know, if it still not working for you.

import pandas as pd
import plotly.graph_objects as go
import numpy as np

def camasRegion():  
    df = pd.read_csv('https://raw.githubusercontent.com/MinCiencia/Datos-COVID19/master/output/producto52/Camas_UCI.csv')
    region = "Ñuble"
    target_region = df.query('Region == @region').drop('Region', axis=1).set_index('Serie').T
    
    chart = []
    for i in target_region:
        if i != 'Camas base (2019)':  #or you can easily drop it from your dataset
            chart  = [go.Scatter(x=target_region.index,y=target_region[i], name=i, mode='lines')]
            
    fig = go.Figure(chart)
    fig.update_layout(title={'text':f'Camas por región: ({region})', 'x':.45},
                      template='plotly_white', hovermode='x',
                      legend_title_text='Estado de cama',
                      xaxis_title="Días",
                      yaxis_title="Cantidad de camas")

    fig.add_annotation(
        x = 1, y = -0.1, 
        text = 'Fuente: Datos obtenidos desde el Ministerio de Ciencia.', 
        showarrow = False, xref='paper', yref='paper', 
        xanchor='right', yanchor='auto', xshift=0, yshift=-20
        )
    fig.show()


camasRegion()

The result: