Find top artists/tracks you've listened to within the past year but not the past 90 days

lastfm_offmyradar.py

# Find top artists/tracks you've listened to within the past year but not the past 90 days
# Before starting, download your Last.fm history from here: https://benjaminbenben.com/lastfm-to-csv/

import datetime
import pandas as pd

# add column names (at time of publishing, the site above didn't do this)
df = pd.read_csv("username.csv",header=0,names=["artist","album","title","timestamp"])
df.timestamp = pd.to_datetime(df.timestamp)
df['trackid'] = df['artist']+df['title']

# create list of tracks played in past 90 days, and between 365 and 90 days ago
last90days = df[df.timestamp > datetime.datetime.now() - pd.to_timedelta("90day")]
last365to90days = df[(df.timestamp < datetime.datetime.now() - pd.to_timedelta("90day")) & (df.timestamp > datetime.datetime.now() - pd.to_timedelta("365day"))]

# create a filter for lapsed ARTISTS
lapsed_artists = last365to90days[~last365to90days['artist'].isin(last90days.artist)]

# create a filter for lapsed ARTIST-TITLE PAIRS
lapsed_titles = last365to90days[~last365to90days['trackid'].isin(last90days['trackid'])]

# results
lapsed_artists['artist'].value_counts().head(50)
lapsed_titles[['artist','title']].value_counts().head(50)