Hello i am new using plotly and I have a question
I am plotting a scatter and a regression line but the points of the scatter cover the line, i want the line to be on top but i donât know how. Here is my code for the plot:
scatter_plot = px.scatter(dfpanda, x=âindexâ, y=âMoistureSensorRawAdcâ,title=âMoistureSensorRawAdcâ)
regression_trace = go.Scatter(x=dfpanda[âindexâ], y=ypred, name=âRegression Lineâ, line=dict(color=âfirebrickâ))
fig = go.Figure()
fig.add_traces(scatter_plot.data)
fig.add_traces(regression_trace)
fig.update_traces(showlegend=True)
fig.update_layout(
title=âMoistureSensorRawAdc con Linea de Regresionâ,
xaxis_title=âIndexâ,
yaxis_title=âMoistureSensorRawAdcâ,
)
And this is how it looks:
Hi @PirateKing Welcome to the forums.
I think you can do so with the latest release of plotly. Check the release notes. I remember reading about it.
If you use go.Scatter() instead of px.Scatter() for the first plot they display in the right order.
e.g. this simplified example:
import plotly.graph_objects as go
import random
x = [random.uniform(0., 2.) for _ in range(20000)]
y = [xval + random.uniform(-1., 1.) for xval in x]
ypred= x
scatter_plot = go.Scatter(x=x, y=y, mode="markers")
regression_trace = go.Scatter(x=x, y=ypred,
line=dict(color='firebrick'))
fig = go.Figure()
fig.add_traces(scatter_plot)
fig.add_traces(regression_trace)
fig.show()
Thereâs further discussion of this at:
2 Likes
Yes, this is better than my suggestion - simpler, and setting zorder>0 on a trace generated by go does position it in front of traces generated by px.
