12.3.10.4.10. Reshaping data

import pandas as pd
import numpy as np


tuples = list(
    zip(
        *[
            ["bar", "bar", "baz", "baz", "foo", "foo", "qux", "qux"],
            ["one", "two", "one", "two", "one", "two", "one", "two"],
        ]
    )
)
index = pd.MultiIndex.from_tuples(tuples, names=["first", "second"])
dataFrame = pd.DataFrame(np.random.randn(8, 2), index=index, columns=["A", "B"])
dataFrame2 = dataFrame[:4]

Stack

stacked = dataFrame2.stack()

stacked.unstack()

		A	B
first	second
bar	one	1.021953	2.347960
	two	-1.133448	-0.915521
baz	one	0.725976	2.472088
	two	0.749382	-0.296489

stacked.unstack(1)

	second	one	two
first
bar	A	1.021953	-1.133448
	B	2.347960	-0.915521
baz	A	0.725976	0.749382
	B	2.472088	-0.296489

stacked.unstack(0)

	first	bar	baz
second
one	A	1.021953	0.725976
	B	2.347960	2.472088
two	A	-1.133448	0.749382
	B	-0.915521	-0.296489

Pivot tables

dataFrame = pd.DataFrame(
    {
        "A": ["one", "one", "two", "three"] * 3,
        "B": ["A", "B", "C"] * 4,
        "C": ["foo", "foo", "foo", "bar", "bar", "bar"] * 2,
        "D": np.random.randn(12),
        "E": np.random.randn(12),
    }
)

pd.pivot_table(dataFrame, values="D", index=["A", "B"], columns=["C"])

	C	bar	foo
A	B
one	A	-0.065709	-1.049577
	B	-0.217735	-0.544323
	C	0.411135	-1.737220
three	A	1.037492	NaN
	B	NaN	-1.840691
	C	-0.516667	NaN
two	A	NaN	0.464021
	B	1.111117	NaN
	C	NaN	1.046043

Time series

indexData = pd.date_range("1/5/2022", periods=100, freq="S")
timeStemps = pd.Series(np.random.randint(0, 500, len(indexData)), index=indexData)
timeStemps.resample("5Min").sum()

2022-01-05    23303
Freq: 5T, dtype: int32

timeStempsUTC = timeStemps.tz_localize("UTC")

timeStempsUTC.tz_convert("US/Eastern")

2022-01-04 19:00:00-05:00    161
2022-01-04 19:00:01-05:00    425
2022-01-04 19:00:02-05:00    216
2022-01-04 19:00:03-05:00    356
2022-01-04 19:00:04-05:00    136
                            ...
2022-01-04 19:01:35-05:00    412
2022-01-04 19:01:36-05:00    154
2022-01-04 19:01:37-05:00     28
2022-01-04 19:01:38-05:00    113
2022-01-04 19:01:39-05:00    100
Freq: S, Length: 100, dtype: int32

ps = timeStemps.to_period()

ps.to_timestamp()

2022-01-05 00:00:00    161
2022-01-05 00:00:01    425
2022-01-05 00:00:02    216
2022-01-05 00:00:03    356
2022-01-05 00:00:04    136
                      ...
2022-01-05 00:01:35    412
2022-01-05 00:01:36    154
2022-01-05 00:01:37     28
2022-01-05 00:01:38    113
2022-01-05 00:01:39    100
Freq: S, Length: 100, dtype: int32

prng = pd.period_range("1990Q1", "2000Q4", freq="Q-NOV")
ts = pd.Series(np.random.randn(len(prng)), prng)
ts.index = (prng.asfreq("M", "e") + 1).asfreq("H", "s") + 9