Created
June 6, 2020 09:00
-
-
Save MarcinMoskala/e1c9d795938fee5542f6571d1441a729 to your computer and use it in GitHub Desktop.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from itertools import count, takewhile | |
def frange(start, stop, step): | |
return takewhile(lambda x: x < stop, count(start, step)) | |
def show_avg_hist(df, column, buckets=40): | |
counts = df[column].value_counts() | |
counts = counts[counts > 20] # Drop rare values | |
min_val = counts.index.min() | |
max_val = counts.index.max() | |
step = (max_val - min_val) / buckets | |
buckets = zip(frange(min_val, max_val + 0.1, step), frange(min_val + step + 0.001, max_val + 0.1 + step, step)) | |
print(column, "from-to\tmean\tdays", sep="\n") | |
s = 0 | |
for (_from, _to) in buckets: | |
chosen = df[column].between(_from, _to) | |
days = sum(chosen) | |
s = s + days | |
print( | |
"{_from:.1f}-{_to:.1f}\t{mean:.1f}\t{days:d}".format(_from=_from, _to=_to, mean=df[chosen]['crimes'].mean(), | |
days=days).replace(".", ",")) | |
def add_day_month_year(df): | |
dates = df["Date"].map(lambda x: x[:10]) | |
df["day_of_month"] = dates.map(lambda x: x.split("/")[1]) | |
df["month"] = dates.map(lambda x: x.split("/")[0]) | |
df["year"] = dates.map(lambda x: x.split("/")[2]) | |
df["date"] = dates | |
return df | |
from pandas import read_csv | |
crimes_df = read_csv("CrimeData.csv") | |
crimes_df = add_day_month_year(crimes_df) | |
crimes_df = crimes_df[crimes_df.year <= "2019"] | |
crimes_df = crimes_df[crimes_df["Primary Type"] == "NARCOTICS"] # "THEFT"... | |
weather_df = read_csv("Weather.csv") | |
weather_df["date"] = weather_df["DATE"].map(lambda x: x[5:7] + "/" + x[8:11] + "/" + x[0:4]) | |
crimes_count_df = crimes_df \ | |
.groupby(["date"]) \ | |
.size() \ | |
.to_frame("crimes") \ | |
.reset_index() | |
df = crimes_count_df.merge(weather_df, on="date", how='left') | |
show_avg_hist(df, column="TMAX", buckets=20) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment