Add connecting line to scatter plot?

[nextgenusfs]

Thanks for your work with plotly_express, it is generally easy to use. We've been trying to add a connecting line to a scatter plot and it doesn't seem like there is a setting to do this? There is the ability to add a trendline smoothed over the data, but we were actually just interested in something simpler. This is possible in plotly by specifying the 'mode' = 'lines+markers'. We were able to come up with a work around, although seems like providing access to the mode in the scatter function would be easier.

Work around is:

import plotly_express as px
iris = px.data.iris()
p = px.scatter(iris, x="sepal_width", y="sepal_length")
for i in range(0, len(p['data'])):
    p['data'][i]['mode'] = 'lines+markers'
p

[nicolaskruchten]

So there are a few approaches you can use here... You could use px.line() or px.scatter() as the basis for your chart, and you could use a few different approached for modifying the mode thereafter. (px.line() sets mode='lines' only by default)

The easiest way to write the code above is:

import plotly_express as px
iris = px.data.iris()
px.scatter(iris, x="sepal_width", y="sepal_length").update_traces(mode="lines+markers")

This works for two reasons:

1. .update_traces() is a single call that does what your for-loop does, much more tersely
2. it also returns the figure after modifying it, so there's no need to assign it, modify it, then return it

We're working on a comprehensive tutorial for how to effectively and easily modify px figures like this :)

[nicolaskruchten]

PS: .update_traces() also accepts magic-underscores, so if you want to modify nested attributes, you can use e.g. .update_traces(marker_line_color="black")

[nextgenusfs]

Fantastic, thanks @nicolaskruchten