[Python3]How to use Seaborn/Matplotlib to graph pandas dataframe

Question

I'm still having troubles to do this

Here is how my data looks like:

     date      positive  negative  neutral
0   2015-09        23         6       18
1   2016-04       709       288      704
2   2016-08      1478       692     1750
3   2016-09      1881       926     2234
4   2016-10      3196      1594     3956

in my csv file I don't have those 0-4 indexes, but only 4 columns from 'date' to 'neutral'. I don't know how to fix my codes to get it look like this

Seaborn code

sns.set(style='darkgrid', context='talk', palette='Dark2')
fig, ax = plt.subplots(figsize=(8, 8))

sns.barplot(x=df['positive'], y=df['negative'], ax=ax)
ax.set_xticklabels(['Negative', 'Neutral', 'Positive'])
ax.set_ylabel("Percentage")
plt.show()

do df.reset_index()

ombk
– ombk

2020-12-04 22:11:31 +00:00
Commented Dec 4, 2020 at 22:11 — ombk
– ombk, Commented Dec 4, 2020 at 22:11

Cameron Riddell · Accepted Answer · 2020-12-04 22:27:15Z

2

To do this in seaborn you'll need to transform your data into long format. You can easily do this via melt:

plotting_df = df.melt(id_vars="date", var_name="sign", value_name="percentage")

print(plotting_df.head())
      date      sign  percentage
0  2015-09  positive          23
1  2016-04  positive         709
2  2016-08  positive        1478
3  2016-09  positive        1881
4  2016-10  positive        3196

Then you can plot this long-format dataframe with seaborn in a straightforward mannter:

sns.set(style='darkgrid', context='talk', palette='Dark2')
fig, ax = plt.subplots(figsize=(8, 8))

sns.barplot(x="date", y="percentage", ax=ax, hue="sign", data=plotting_df)

answered Dec 4, 2020 at 22:27

Cameron Riddell

13.8k14 silver badges21 bronze badges

Sign up to request clarification or add additional context in comments.

1 Comment

ombk Over a year ago

yep indeed this is also a great way!

ombk · Accepted Answer · 2020-12-04 22:18:45Z

1

Based on the data you posted

sns.set(style='darkgrid', context='talk', palette='Dark2')
# fig, ax = plt.subplots(figsize=(8, 8))

df.plot(x="date",y=["positive","neutral","negative"],kind="bar")
plt.xticks(rotation=-360)
# ax.set_xticklabels(['Negative', 'Neutral', 'Positive'])
# ax.set_ylabel("Percentage")
plt.show()

answered Dec 4, 2020 at 22:18

ombk

2,1091 gold badge6 silver badges16 bronze badges

2 Comments

ombk Over a year ago

You can do it in seaborn, however, you will need to change how your data is structured

ombk Over a year ago

the dates should print vertically if not. plt.xticks(rotation=90) keep rotating until u get the right direction @Elainayg

Collectives™ on Stack Overflow

[Python3]How to use Seaborn/Matplotlib to graph pandas dataframe

2 Answers 2

1 Comment

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

2 Comments

Your Answer

Sign up or log in

Post as a guest