by_weekday.py

from pandas import read_csv
crimes_df = read_csv("CrimeData.csv")
crimes_df = crimes_df[crimes_df.year <= "2019"]
days_df = read_csv("DaysData.csv")


from datetime import datetime
def to_weekday(date):
    return datetime.strptime(date, "%m/%d/%Y").strftime("%A")

crimes_df["weekday"] = crimes_df.date.map(to_weekday)

crimes_count_df = crimes_df \
    .groupby(["weekday"]) \
    .size() \
    .to_frame("crimes") \
    .reset_index()

print(crimes_count_df.to_string())