Add line plots to Facetgrid plot

Question

I plotted a simple RelPlotusing Seaborn, which returned me a Facetgrid object. The code I used is the following:

import seaborn as sns

palette = sns.color_palette("rocket_r")

g1 = sns.relplot(
    data=df,
    x="number_of_weeks", y="avg_streams",
    hue="state", col="state",
    kind="line", palette=palette,
    height=5, aspect=1, facet_kws=dict(sharex=False), col_wrap=3, 
)

Which shows the following plot:

And the Dataframe I'm using has the following structure:

   avg_streams  date    year number_of_weeks state
4   0.104011    31-01   2020    4   it
5   1.211951    07-02   2020    5   it
6   0.559374    14-02   2020    6   it
7   0.304257    21-02   2020    7   it
8   0.199218    28-02   2020    8   it
... ... ... ... ... ...
175 -0.938890   26-06   2020    25  br
176 -0.483821   03-07   2020    26  br
177 -0.083704   10-07   2020    27  br
178 0.165312    17-07   2020    28  br
179 0.218601    24-07   2020    29  br

I would like to add other lineplots to the single subplot. My final goal is to plot all the lines in each single subplot, but highlight a different state for each different subplot.

So I would like to get something like this for each different subplot in my Facetgrid:

And this is the code I wrote for the previous plot:

import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np

palette = {c:'red' if c=='it' else 'grey' for c in df.state.unique()}

fig, ax = plt.subplots(figsize=(15, 7))    
plot=sns.lineplot(ax=ax,x="number_of_weeks", y="avg_streams", hue="state", data=df, palette=palette)

lines = ax.get_lines()
lines[0].set_linewidth(5)

plot.set(title='Streams trend')

But I can't figure out how to "merge" the 2 plots. How can I achieve my goal?

EDIT: I tried to add the plot "manually" selecting the single axes of my Facetgrid. I followed this question: Add lineplot to subplot and I was able to add a simple line.

This is my try, I tried to add a simple line to my already existing plot:

palette = sns.color_palette("rocket_r")

g1 = sns.relplot(
    data=df,
    x="number_of_weeks", y="avg_streams",
    hue="state", col="state",
    kind="line", palette=palette,
    height=5, aspect=1, facet_kws=dict(sharex=False), col_wrap=3, 
)
axes = g1.fig.axes
print(axes)

axes[0].plot([20, 30], [40, 50], 'k-')

put all of your data in the same dataframe, make as many calls to `g.map` as you need — Paul H, Sep 18 '20 at 17:46
@PaulH What do you mean? All my data is already in the same dataframe — Mattia Surricchio, Sep 18 '20 at 17:48
I think this question is exactly [this example](http://seaborn.pydata.org/examples/timeseries_facets.html) from the documentation? — mwaskom, Sep 18 '20 at 19:47
@mwaskom Oh, thank you! That's exactly what I needed, although in the end I managed to do it on my own. — Mattia Surricchio, Sep 18 '20 at 19:57

score 5 · Accepted Answer · answered Sep 18 '20 at 19:24

I found a solution, the code is the following:

import seaborn as sns

palette = sns.color_palette("tab10",6)

g1 = sns.relplot(
    data=df,
    x="number_of_weeks", y="avg_streams",
    hue="state", col="state",
    kind="line", palette=palette,
    height=5, aspect=1, facet_kws=dict(sharex=False), col_wrap=3, linewidth=5, zorder= 5
)
axes = g1.fig.axes
states = df.state.unique()

for index, state in enumerate (states, start=0):
    df_temp = df.loc[df['state'] != state]
    palette1 = {c:'grey' for c in df_temp.state.unique()}
    sns.lineplot(ax=axes[index],x="number_of_weeks", y="avg_streams", hue="state", data=df_temp, palette=palette1)
    axes[index].get_legend().remove()
    axes[index].set_xlabel('')

And this is the result:

Add line plots to Facetgrid plot

1 Answers1