How to use mapped values for numerical chart axis ticks

Say I have a variable, df, that contains some data that looks like this in pandas

Date          Number Letter
2017-10-31    2      B
2019-08-31    1      A
2021-11-30    3      C
...

I'd like to chart this out in seaborn, so I use some code like this

sb.lineplot(data=df, x=Date, y=Number)
plt.show()

And I get an orderly line chart that looks something like this

However, i'd like to have the y axis tick labels to preserve the numerical order of the number column, but present as labels from the letter column. For instance, instead of the y axis being 10 - 17 as above, i'd like the y axis to read j - t.

Is there a way to implement that through the sb.lineplot arguments, or would I need another way?

IIUC, you can use FixedLocator and FixedFormatter:

import matplotlib.ticker as ticker

ax = sb.lineplot(data=df, x='Date', y='Number')
ax.yaxis.set_major_locator(ticker.FixedLocator(df['Number']))
ax.yaxis.set_major_formatter(ticker.FixedFormatter(df['Letter']))
plt.show()

Output:

• It's more straightforward to plot 'Letter' on the y-axis.
• If the values on the y-axis require a specific order, then set the 'Letter' column with pd.Categorical.
  • categories=sorted(df.Letter, reverse=True), set reverse=False to have A at the top of the y-axis.
• Tested in python 3.11.3, pandas 2.0.2, matplotlib 3.7.1, seaborn 0.12.2

import pnadas as pd
import seaborn as sns  # this is the standard alias
import matplotlib.pyplot as plt

# sample data
data = {'Date': ['2017-10-31', '2019-08-31', '2021-11-30'], 'Number': [2, 1, 3], 'Letter': ['B', 'A', 'C']}
df = pd.DataFrame(data)

# confirm the Date column is a datetime date dtype
df.Date = pd.to_datetime(df.Date).dt.date

# order the Letter column
df.Letter = pd.Categorical(df.Letter, categories=sorted(df.Letter, reverse=True))

# plot
fig = plt.figure(figsize=(10, 6))
ax = sns.lineplot(data=df, x='Date', y='Letter')