Python 3: pandas group multiindex and mean of 2 separate columns per multiindex

Question

What code do I need to perform a mean of both drugs and health per session by type.

df[['drugs', 'health', 'session', 'type']]

these all are currently columns. I though it ought to be indexed by both session (first) and then type .

grouped_S = grouped['session'].as_index=True
grouped_c = grouped['type'].as_index=True
grouped_dh = grouped['drugs', 'health'].mean().as_index=False

How do I correct this code to make it work: getting an average of drugs and health per session and by type?

Answer 1

IIUC, try this:

df.groupby(['session', 'type'])[['drugs', 'health']].mean()