plotting_pivots_ploty.md

How to plot a Pandas Pivot table with plotly

To plot the pandas pivot table pivot_table using Plotly Express, you can follow these steps:

import pandas as pd
import plotly.express as px

pivot_table = df.pivot_table(
    index="day_dt", columns="pubid", values="total_impressions", aggfunc="sum"
)

df_long = pivot_table.reset_index().melt(id_vars="day_dt", var_name="pubid", value_name="total_impressions")

fig = px.line(df_long, x="day_dt", y="total_impressions", color="pubid")

fig.show()

In this example, we first reset the index of pivot_table using the reset_index() method. Then, we use the melt() function to reshape the DataFrame to a long format, similar to what was done in the previous example.

After that, we create the line graph using px.line. We specify the DataFrame df_long as the input, set the x-axis values to "day_dt", the y-axis values to "total_impressions", and the color encoding to "thirdpartypubid".

Finally, we display the graph using fig.show().