Tricks

import pandas as pd

Explode

also new for pandas 0.25

List

If have a list within a column and you want to create a row for each item in that list.

df = pd.DataFrame({
    'article': ['a', 'b', 'c'],
    'tags': [['happy', 'fun'], ['sad', 'frustrating', 'stressful'], ['ಠ_ಠ']],
    'views': [100, 10, 10000]
})

df

	article	tags	views
0	a	[happy, fun]	100
1	b	[sad, frustrating, stressful]	10
2	c	[ಠ_ಠ]	10000

df.explode('tags')

	article	tags	views
0	a	happy	100
0	a	fun	100
1	b	sad	10
1	b	frustrating	10
1	b	stressful	10
2	c	ಠ_ಠ	10000

Comma-seperated

If have comma seperated values within a column and you want to create a row for each item in the string

df = pd.DataFrame({
    'article': ['a', 'b', 'c'],
    'tags': ['happy,fun', 'sad,frustrating,stressful', 'ಠ_ಠ'],
    'views': [100, 10, 10000]
})

df

	article	tags	views
0	a	happy,fun	100
1	b	sad,frustrating,stressful	10
2	c	ಠ_ಠ	10000

(
    df
    .assign(tags=df.tags.str.split(","))
    .explode('tags')
)

	article	tags	views
0	a	happy	100
0	a	fun	100
1	b	sad	10
1	b	frustrating	10
1	b	stressful	10
2	c	ಠ_ಠ	10000

Query

for easy to read sql-like slices, good for chaining

(
    df
    .query('views > 10 & views < 10000')
)

	article	tags	views
0	a	happy,fun	100

can also pass variables with @

threshold = 100

(
    df
    .query('views == @threshold')
)

	article	tags	views
0	a	happy,fun	100

Multiple Conditions

Sometimes we want to filter our dataframe using a list

some_list = ['a', 'b']

df[df.article.isin(some_list)]

	article	tags	views
0	a	happy,fun	100
1	b	sad,frustrating,stressful	10

we can reverse this easily with ~

df[~df.article.isin(some_list)]

	article	tags	views
2	c	ಠ_ಠ	10000

Assign

Skipping basics since you can just look here: https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.DataFrame.assign.html

Create columns with variable names, also f strings (prob cleaner to use than {}.format()

col_name = 'cat'

(
    df
    .assign(
        **{f'some_{col_name}': ['haku', 'nagi', 'poki']}
    )
    .style.set_caption("some random dataframe I imagined")
)

some random dataframe I imagined
	article	tags	views	some_cat
0	a	happy,fun	100	haku
1	b	sad,frustrating,stressful	10	nagi
2	c	ಠ_ಠ	10000	poki

can also set a caption for the dataframe lol