In [1]:

import numpy as np
import pandas as pd

pd.set_option('display.float_format', '{:,.1f}'.format)

In [2]:

dat = pd.read_csv('data/weather-6m.csv')

Exercise:

Calculate the average air_temp by month.

In [3]:

grp = dat.groupby('month')

In [4]:

grp['air_temp'].mean()

Out[4]:

month
1   -10.0
2    -3.0
3     2.1
4     7.0
5    14.0
6    18.1
Name: air_temp, dtype: float64

Exercise:

Compute summary statistics on air_temp and dew_point using the describe method.

In [5]:

grp[['air_temp', 'dew_point']].describe()

Out[5]:

	air_temp								dew_point
	count	mean	std	min	25%	50%	75%	max	count	mean	std	min	25%	50%	75%	max
month
1	712.0	-10.0	6.2	-29.4	-13.3	-10.0	-5.6	2.8	712.0	-14.1	6.8	-32.8	-18.3	-13.9	-8.9	1.0
2	644.0	-3.0	6.8	-19.4	-7.2	-2.2	1.7	15.0	644.0	-7.3	7.3	-22.8	-12.2	-7.2	-2.2	8.3
3	713.0	2.1	6.7	-13.3	-1.7	2.2	5.6	22.8	713.0	-3.4	6.1	-17.2	-7.8	-2.8	0.6	13.3
4	691.0	7.0	6.0	-2.8	2.8	5.6	9.4	28.3	691.0	0.3	6.1	-13.3	-3.9	-1.1	3.9	16.7
5	713.0	14.0	5.1	1.1	10.6	13.9	17.2	28.3	713.0	6.2	4.6	-6.1	2.8	6.7	10.0	18.3
6	688.0	18.1	6.0	3.3	13.8	17.8	22.8	33.3	688.0	12.3	5.5	-3.3	8.9	11.7	17.2	23.3

Exercise:

For January and February and 0 - 11 hours, calculate the average and standard deviation of air_temp grouping by month and hour of the day. Name your result columns air_temp_mean and air_temp_sd.

Your result DataFrame should have 24 rows, the number of months (2) times the number of hours (12).

$2 * 12 = 24$

In [6]:

idx = dat.month.isin([1, 2]) & (dat.hour < 12)
grp2 = dat[idx].groupby(['month', 'hour'])
hourly_temp = grp2.agg(
    air_temp_mean=('air_temp', 'mean'),
    air_temp_sd=('air_temp', 'std')
)

Exercise:

By month, calculate quantiles for air_temp using the quantiles defined in breaks.

Hint: Use the quantile method defined on a Series (pd.Series.quantile).

In [7]:

breaks = [0.01, 0.25, 0.5, 0.75, 0.99]

In [8]:

grp3 = dat.groupby('month')
grp3.apply(lambda x: x.air_temp.quantile(breaks))

Out[8]:

air_temp	0.0	0.2	0.5	0.8	1.0
month
1	-25.0	-13.3	-10.0	-5.6	1.1
2	-18.3	-7.2	-2.2	1.7	12.8
3	-11.0	-1.7	2.2	5.6	19.3
4	-2.2	2.8	5.6	9.4	23.4
5	2.8	10.6	13.9	17.2	27.8
6	6.1	13.8	17.8	22.8	32.2