How to have a datetime slider like in px.scatter for go.Scattergl from plotly-CodePudding

Morning everyone,

My code is :

fig = go.Figure()
fig.add_trace(
    go.Scattergl(x=result2['DateTime'], y=result2['Voc (V)'], mode='markers'))
# Set title
fig.update_layout(
    title_text="VOC (V) vs Time", xaxis_title="Time", yaxis_title="VOC (V)")
# Add range slider
fig.update_layout(
    xaxis=dict(
        rangeselector=dict(
            buttons=list([
                dict(count=1,
                     label="1m",
                     step="month",
                     stepmode="backward"),
                dict(count=6,
                     label="6m",
                     step="month",
                     stepmode="backward"),
                dict(count=1,
                     label="YTD",
                     step="year",
                     stepmode="todate"),
                dict(count=1,
                     label="1y",
                     step="year",
                     stepmode="backward"),
                dict(step="all")
            ])
        ),
        rangeslider=dict(
            visible=True
        ),
        type="date"
    )
)
from plotly import offline
offline.plot(fig)

It gives me this image as an output :

And this code :

import plotly.express as px

df = result2
fig = px.scatter(df, x="DateTime", y="Voc (V)",
                 color="Eclairement (W/m²)")
fig["layout"].pop("updatemenus")
from plotly import offline
offline.plot(fig)

And i get this image as output :

My question : How can i have the Slider the 3rd axis color (which is coloring y axis = Voc (V))

PS : result2 is my dataframe of some columns.

CodePudding user response：

this is a workaround. The rangeslider visual does not display for a scattergl trace
have simulated some data to demonstrate
assuming scattergl is being used as it performs better for large number of points. Concept
- use subset of points for a scatter trace so that it does not use as many resources
- this second trace uses a different yaxis so that it can be hidden by setting domain to be very small
- based on number of points you may want to choose a large number for nth observation for rangeslider visual

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

df = pd.DataFrame(
    {
        "DateTime": pd.date_range("1-aug-2021", freq="1H", periods=1000),
        "Voc (V)": ((np.sin(np.tile(np.linspace(-np.pi, np.pi, 100), 10))   1) * 700)
        * np.random.uniform(0.8, 0.9, 1000),
        "Eclairement (W/m²)": (np.sin(np.tile(np.linspace(-np.pi, np.pi, 100), 10))   1)
        * 5,
    }
)

# every nth row for rangeslider
df2 = df.iloc[::10, :]

fig = go.Figure(
    [
        go.Scattergl(
            x=df["DateTime"],
            y=df["Voc (V)"],
            mode="markers",
            marker_color=df["Eclairement (W/m²)"],
            name="main",
        ),
        go.Scatter(
            x=df2["DateTime"],
            y=df2["Voc (V)"],
            mode="markers",
            marker_color=df2["Eclairement (W/m²)"],
            name="rangeslider",
            yaxis="y2",
        ),
    ]
)

fig.update_layout(
    xaxis_rangeslider_visible=True,
    yaxis2={"domain": [0, 0.001], "visible": False},
    showlegend=False,
)