Curso DS - pandas

import numpy as np import pandas as pd from numpy.random import randn

np.random.seed(101)

df = pd.DataFrame(randn(5,4), ["a", "b", "c", "d", "e"], ["one", "two", "three", "four"]) df

onefloat64

twofloat64

a

2.706849839

0.6281327088

b

0.6511179479

-0.3193180446

c

-2.018168244

0.7401220571

d

0.1886953094

-0.7588720562

e

0.1907943224

1.978757324

randn(5,4)

# Call columns # df.one df[["one", "two"]]

onefloat64

twofloat64

a

2.706849839

0.6281327088

b

0.6511179479

-0.3193180446

c

-2.018168244

0.7401220571

d

0.1886953094

-0.7588720562

e

0.1907943224

1.978757324

# Create new coloum df["new_column"] = df["one"] + df["two"]

df

onefloat64

twofloat64

a

2.706849839

0.6281327088

b

0.6511179479

-0.3193180446

c

-2.018168244

0.7401220571

d

0.1886953094

-0.7588720562

e

0.1907943224

1.978757324

df.drop("new_column", axis=1)

onefloat64

twofloat64

a

2.706849839

0.6281327088

b

0.6511179479

-0.3193180446

c

-2.018168244

0.7401220571

d

0.1886953094

-0.7588720562

e

0.1907943224

1.978757324

df

onefloat64

twofloat64

a

2.706849839

0.6281327088

b

0.6511179479

-0.3193180446

c

-2.018168244

0.7401220571

d

0.1886953094

-0.7588720562

e

0.1907943224

1.978757324

#Para eliminarlo del df original. Inplace true df.drop("new_column", axis=1, inplace=True) df

onefloat64

twofloat64

a

2.706849839

0.6281327088

b

0.6511179479

-0.3193180446

c

-2.018168244

0.7401220571

d

0.1886953094

-0.7588720562

e

0.1907943224

1.978757324

df >= 0

onebool

twobool

a

true

b

true

false

c

false

true

d

true

false

e

true

df[df >= 0]

onefloat64

twofloat64

a

2.706849839

0.6281327088

b

0.6511179479

nan

c

nan

0.7401220571

d

0.1886953094

nan

e

0.1907943224

1.978757324

df["one"]>= 0

# Si quiero filtrar df[df["one"]>=0]

onefloat64

twofloat64

a

2.706849839

0.6281327088

b

0.6511179479

-0.3193180446

d

0.1886953094

-0.7588720562

e

0.1907943224

1.978757324

# Si negar una condición df[~(df["one"]>=0)]

onefloat64

twofloat64

c

-2.018168244

0.7401220571

This chart is empty

Chart was probably not set up properly in the notebook

df[(df["one"]>=0) & (df["four"]< 1)]

onefloat64

twofloat64

a

2.706849839

0.6281327088

b

0.6511179479

-0.3193180446

d

0.1886953094

-0.7588720562

e

0.1907943224

1.978757324

# Quitar los index df.reset_index()

indexobject

onefloat64

0

a

2.706849839

1

b

0.6511179479

2

c

-2.018168244

3

d

0.1886953094

4

e

0.1907943224

index = ["a1", "a2", "a3", "a4", "a5"] df["index"] = index df.set_index("index")

onefloat64

twofloat64

a1

2.706849839

0.6281327088

a2

0.6511179479

-0.3193180446

a3

-2.018168244

0.7401220571

a4

0.1886953094

-0.7588720562

a5

0.1907943224

1.978757324

# Index Levels outside = ['G1','G1','G1','G2','G2','G2'] inside = [1,2,3,1,2,3] hier_index = list(zip(outside,inside)) hier_index = pd.MultiIndex.from_tuples(hier_index)

type(hier_index)

hier_index

df = pd.DataFrame(np.random.randn(6,2),index=hier_index,columns=['A','B']) df

Afloat64

Bfloat64

('G1', 1)

-0.9932635

0.196799505

('G1', 2)

-1.136644594

0.0003664796056

('G1', 3)

1.025984152

-0.1565979043

('G2', 1)

-0.03157914391

0.6498258335

('G2', 2)

2.154846443

-0.6102588558

('G2', 3)

-0.75532534

-0.3464185035

df.loc["G1", 1]

df.index.names df.index.names = ["Grupos", "numeros"]

df.index

df.xs(1, level="numeros")

Afloat64

Bfloat64

G1

-0.9932635

0.196799505

G2

-0.03157914391

0.6498258335

df = pd.DataFrame(np.random.rand(5,4)) df[df <= 0.5] = np.nan df

0float64

1float64

0

nan

0.6376586528

1

0.6517677035

nan

2

0.6599245189

0.5296233988

3

0.78452185

0.6872420376

4

0.975361127

nan

df

0float64

1float64

0

nan

0.6376586528

1

0.6517677035

nan

2

0.6599245189

0.5296233988

3

0.78452185

0.6872420376

4

0.975361127

nan

# Retorna los row que tenga almenos 3 not-nan df.dropna(thresh=3)

0float64

1float64

0

nan

0.6376586528

2

0.6599245189

0.5296233988

3

0.78452185

0.6872420376

df.fillna(0)

0float64

1float64

0

0.6376586528

1

0.6517677035

0

2

0.6599245189

0.5296233988

3

0.78452185

0.6872420376

4

0.975361127

0

df.mean()

data = {'Company':['GOOG','GOOG','MSFT','MSFT','FB','FB'], 'Person':['Sam','Charlie','Amy','Vanessa','Carl','Sarah'], 'Sales':[200,120,340,124,243,350]} df = pd.DataFrame(data) df

Companyobject

Personobject

0

GOOG

Sam

1

GOOG

Charlie

2

MSFT

Amy

3

MSFT

Vanessa

4

FB

Carl

5

FB

Sarah

df_company = df.groupby("Company") df_company.mean()

Salesfloat64

FB

296.5

GOOG

160

MSFT

232

df_company.sum()

Salesint64

FB

593

GOOG

320

MSFT

464

df_company.sample()

Companyobject

Personobject

5

FB

Sarah

0

GOOG

Sam

2

MSFT

Amy

df.groupby("Company").sum().loc["FB"]

df.groupby("Company").count()

Personint64

Salesint64

FB

2

GOOG

2

MSFT

2

df.groupby("Company").describe().transpose()

FBfloat64

GOOGfloat64

('Sales', 'count')

2

('Sales', 'mean')

296.5

160

('Sales', 'std')

75.66042559

56.56854249

('Sales', 'min')

243

120

('Sales', '25%')

269.75

140

('Sales', '50%')

296.5

160

('Sales', '75%')

323.25

180

('Sales', 'max')

350

200

df_books = pd.read_csv("/work/bestsellers-with-categories_e591527f-ae45-4fa5-b0d1-d50142128fa6.csv") df_books

Nameobject

Publication Manual of the American Psychological Association, 6th Edition1.8%

StrengthsFinder 2.01.6%

349 others96.5%

Authorobject

Jeff Kinney2.2%

Suzanne Collins2%

246 others95.8%

0

10-Day Green Smoothie Cleanse

JJ Smith

1

11/22/63: A Novel

Stephen King

2

12 Rules for Life: An Antidote to Chaos

Jordan B. Peterson

3

1984 (Signet Classics)

George Orwell

4

5,000 Awesome Facts (About Everything!) (National Geographic Kids)

National Geographic Kids

5

A Dance with Dragons (A Song of Ice and Fire)

George R. R. Martin

6

A Game of Thrones / A Clash of Kings / A Storm of Swords / A Feast of Crows / A Dance with Dragons

George R. R. Martin

7

A Gentleman in Moscow: A Novel

Amor Towles

8

A Higher Loyalty: Truth, Lies, and Leadership

James Comey

9

A Man Called Ove: A Novel

Fredrik Backman

df_books.groupby("Author").sum()

User Ratingfloat64

4.0 - 57.599999999999994

Reviewsint64

257 - 278329

Abraham Verghese

9.2

9732

Adam Gasiewski

4.4

3113

Adam Mansbach

4.8

9568

Adir Levy

4.8

8170

Admiral William H. McRaven

4.7

10199

Adult Coloring Book Designs

4.5

2313

Alan Moore

4.8

3829

Alex Michaelides

4.5

27536

Alice Schertle

4.9

1884

Allie Brosh

4.7

4896

df_books.groupby("Author").agg(["min", "max"])

Name minobject

Cutting for Stone0.4%

Milk and Vine: Inspirational Quotes From Classic Vines0.4%

246 others99.2%

Name maxobject

Cutting for Stone0.4%

Milk and Vine: Inspirational Quotes From Classic Vines0.4%

246 others99.2%

Abraham Verghese

Cutting for Stone

Adam Gasiewski

Milk and Vine: Inspirational Quotes From Classic Vines

Adam Mansbach

Go the F**k to Sleep

Adir Levy

What Should Danny Do? (The Power to Choose Series)

Admiral William H. McRaven

Make Your Bed: Little Things That Can Change Your Life...And Maybe the World

Adult Coloring Book Designs

Adult Coloring Book Designs: Stress Relief Coloring Book: Garden Designs, Mandalas, Animals, and Paisley Patterns

Alan Moore

Watchmen

Alex Michaelides

The Silent Patient

Alice Schertle

Little Blue Truck

Allie Brosh

Hyperbole and a Half: Unfortunate Situations, Flawed Coping Mechanisms, Mayhem, and Other Things That Happened

df_books.groupby("Author").agg({"Year": "mean", "User Rating": "min"})

Yearfloat64

2009.0 - 2019.0

User Ratingfloat64

3.3 - 4.9

Abraham Verghese

2010.5

4.6

Adam Gasiewski

2017

4.4

Adam Mansbach

2011

4.8

Adir Levy

2019

4.8

Admiral William H. McRaven

2017

4.7

Adult Coloring Book Designs

2016

4.5

Alan Moore

2009

4.8

Alex Michaelides

2019

4.5

Alice Schertle

2014

4.9

Allie Brosh

2013

4.7

df_books.groupby('Author').agg({"Year": lambda x: x.mean() - 2000})

Yearfloat64

9.0 - 19.0

Abraham Verghese

10.5

Adam Gasiewski

17

Adam Mansbach

11

Adir Levy

19

Admiral William H. McRaven

17

Adult Coloring Book Designs

16

Alan Moore

9

Alex Michaelides

19

Alice Schertle

14

Allie Brosh

13

url = "https://static.platzi.com/media/public/uploads/hpcharactersdataraw_3d934e85-dfa4-42ec-8520-fadfbecae574.json" pd.read_json(url)

Nameobject

Cornelius Agrippa0.2%

Aunt Marge Dursley0.2%

1348 others99.7%

Linkobject

https://www.hp-lexicon.org/character/cornelius-agrippa/0.2%

https://www.hp-lexicon.org/character/dursley-family/marjorie-dursley/0.2%

1348 others99.7%

0

Mrs. Abbott

https://www.hp-lexicon.org/character/abbott-family/mrs-abbott/

1

Hannah Abbott

https://www.hp-lexicon.org/character/abbott-family/hannah-abbott/

2

Abel Treetops

https://www.hp-lexicon.org/character/abel-treetops/

3

Euan Abercrombie

https://www.hp-lexicon.org/character/abercrombie-euan/

4

Aberforth Dumbledore

https://www.hp-lexicon.org/character/dumbledore-family/aberforth-dumbledore/

5

Abernathy

https://www.hp-lexicon.org/character/abernathy/

6

Abraham Peasegood

https://www.hp-lexicon.org/character/abraham-peasegood/

7

Abraham Potter

https://www.hp-lexicon.org/character/potter-family/abraham-potter/

8

Abraxas Malfoy

https://www.hp-lexicon.org/character/malfoy-family/abraxas-malfoy/

9

Achilles Tolliver

https://www.hp-lexicon.org/character/achilles-tolliver/

left = pd.DataFrame({'key': ['K0', 'K1', 'K2', 'K3'], 'A': ['A0', 'A1', 'A2', 'A3'], 'B': ['B0', 'B1', 'B2', 'B3']}) right = pd.DataFrame({'key': ['K0', 'K1', 'K2', 'K3'], 'C': ['C0', 'C1', 'C2', 'C3'], 'D': ['D0', 'D1', 'D2', 'D3']}) print(left, "\n\n", right)

  key   A   B
0  K0  A0  B0
1  K1  A1  B1
2  K2  A2  B2
3  K3  A3  B3 

   key   C   D
0  K0  C0  D0
1  K1  C1  D1
2  K2  C2  D2
3  K3  C3  D3

pd.merge(left, right, how="inner", on="key")

keyobject

Aobject

0

K0

A0

1

K1

A1

2

K2

A2

3

K3

A3

left = pd.DataFrame({ 'A': ['A0', 'A1', 'A2', 'A3'], 'B': ['B0', 'B1', 'B2', 'B3']}, index = ['K0', 'K1', 'K2', 'K3']) right = pd.DataFrame({ 'C': ['C0', 'C1', 'C2', 'C3'], 'D': ['D0', 'D1', 'D2', 'D3']}, index = ['K0', 'K1', 'K2', 'K3']) print(left, "\n\n", right)

     A   B
K0  A0  B0
K1  A1  B1
K2  A2  B2
K3  A3  B3 

      C   D
K0  C0  D0
K1  C1  D1
K2  C2  D2
K3  C3  D3

left.join(right)

Aobject

Bobject

K0

A0

B0

K1

A1

B1

K2

A2

B2

K3

A3

B3

df = pd.DataFrame({'col1':[1,2,3,4],'col2':[444,555,666,444],'col3':['abc','def','ghi','xyz']}) df.head()

col1int64

col2int64

0

1

444

1

2

555

2

3

666

3

4

444

df["col2"].unique()

# Si quiero la cuenta de los unicos df["col2"].nunique()

# Retorna el numero de ocurencias de cada label df["col2"].value_counts()

df

col1int64

col2int64

0

1

444

1

2

555

2

3

666

3

4

444

df["col2"].apply(lambda x: x*2)

df.drop("col1", axis=1)

col2int64

col3object

0

444

abc

1

555

def

2

666

ghi

3

444

xyz

#sort by columns df.sort_values('col2')

col1int64

col2int64

0

1

444

3

4

444

1

2

555

2

3

666

data = {'A':['foo','foo','foo','bar','bar','bar'], 'B':['one','one','two','two','one','one'], 'C':['x','y','x','y','x','y'], 'D':[1,3,2,5,4,1]} df = pd.DataFrame(data) df

Aobject

Bobject

0

foo

one

1

foo

one

2

foo

two

3

bar

two

4

bar

one

5

bar

one

df.pivot_table(values="D", columns="A").T

Dfloat64

bar

3.333333333

foo

2

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

bar onefloat64

bar twofloat64

x

4

nan

y

1

5