Visualización de dataframes, selección y asignación de datos

Jeshua Romero Guadarrama
39 Min para leer
09 Aug, 2021
- Python
- Pandas

Índice de contenido

En primer lugar, haciendo uso de los conceptos aprendidos en la anterior lección, vamos a cargar unos datos en un DataFrame a partir de un archivo descargado del repositorio UCI de Machine Learning: https://archive.ics.uci.edu/ml/datasets/Adult

python

import pandas as pd
import numpy as np

python

file_path = "C:/Users/user/Desktop/adult.data"

col_names = ["age", 
            "workclass",
            "fnlwgt", 
            "education",
            "education-num",
            "marital-status",
            "occupation",
            "relationship",
            "race",
            "sex",
            "capital-gain",
            "capital-loss",
            "hours-per-week",
            "native-country",
            "income"]


df = pd.read_csv(filepath_or_buffer = file_path,
           header = 0,
           names = col_names)
df

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
0	50	Self-emp-not-inc	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	0	40	United-States	<=50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32555	27	Private	257302	Assoc-acdm	12	Married-civ-spouse	Tech-support	Wife	White	Female	0	0	38	United-States	<=50K
32556	40	Private	154374	HS-grad	9	Married-civ-spouse	Machine-op-inspct	Husband	White	Male	0	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	9	Widowed	Adm-clerical	Unmarried	White	Female	0	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	9	Married-civ-spouse	Exec-managerial	Wife	White	Female	15024	0	40	United-States	>50K

32560 rows × 15 columns

Visualización básica de DataFrames

df.head(): primeras filas

python

# df.head() nos muestra, por defecto, las primeras 5 filas del dataframe
df.head()

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	hours-per-week	native-country	income
0	50	Self-emp-not-inc	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	40	United-States	<=50K

python

# Si quieremos ver más o menos de 5 filas, como por ejemplo 2, podemos especificarlo de la siguiente manera:
df.head(n=2)

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
0	50	Self-emp-not-inc	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	0	40	United-States	<=50K

df.tail(): últimas filas

Funciona de forma idéntica a df.head(), pero mostrando las últimas filas en lugar de las primeras:

python

df.tail()

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	hours-per-week	native-country	income
32555	27	Private	257302	Assoc-acdm	12	Married-civ-spouse	Tech-support	Wife	White	Female	0	38	United-States	<=50K
32556	40	Private	154374	HS-grad	9	Married-civ-spouse	Machine-op-inspct	Husband	White	Male	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	9	Widowed	Adm-clerical	Unmarried	White	Female	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	9	Married-civ-spouse	Exec-managerial	Wife	White	Female	15024	40	United-States	>50K

python

df.tail(n=2)

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	9	Married-civ-spouse	Exec-managerial	Wife	White	Female	15024	0	40	United-States	>50K

df.sample(): filas aleatorias

En este caso, por defecto solo se muestra una fila aleatoria, aunque también se puede especificar cualquier otro número para mostrar más filas.

python

df.sample()

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
1941	24	Local-gov	249101	HS-grad	9	Divorced	Protective-serv	Unmarried	Black	Female	0	0	40	United-States	<=50K

python

df.sample(n = 5)

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-loss	hours-per-week	native-country	income
32037	52	Private	294991	Bachelors	13	Married-civ-spouse	Prof-specialty	Husband	White	Male	0	40	United-States	>50K
3507	37	Private	265038	Some-college	10	Married-civ-spouse	Transport-moving	Husband	White	Male	0	50	United-States	<=50K
807	64	Private	270333	Bachelors	13	Married-civ-spouse	Prof-specialty	Husband	White	Male	0	40	United-States	>50K
17249	25	Private	456618	7th-8th	4	Never-married	Machine-op-inspct	Unmarried	White	Male	0	40	El-Salvador	<=50K
17770	54	Private	93605	HS-grad	9	Married-civ-spouse	Sales	Husband	White	Male	1848	40	United-States	>50K

python

df.sample(frac= 0.1)

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
22204	38	Private	201454	Some-college	10	Divorced	Adm-clerical	Unmarried	White	Female	0	0	40	United-States	<=50K
522	37	Local-gov	186035	Some-college	10	Married-civ-spouse	Tech-support	Husband	White	Male	0	0	45	United-States	>50K
22175	22	Private	137591	Some-college	10	Never-married	Sales	Own-child	White	Male	0	0	35	United-States	<=50K
13421	53	Private	366957	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	Asian-Pac-Islander	Male	99999	0	50	India	>50K
28741	41	Private	118721	12th	8	Divorced	Adm-clerical	Not-in-family	Amer-Indian-Eskimo	Male	0	0	40	United-States	<=50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
26171	59	Self-emp-inc	141326	Assoc-voc	11	Divorced	Prof-specialty	Not-in-family	White	Male	0	0	50	United-States	>50K
19613	46	Private	194431	HS-grad	9	Never-married	Tech-support	Other-relative	White	Male	0	0	40	United-States	<=50K
17313	76	?	84755	Some-college	10	Widowed	?	Unmarried	White	Female	0	0	40	United-States	<=50K
13896	24	Private	193920	Masters	14	Never-married	Priv-house-serv	Not-in-family	White	Female	0	0	45	?	<=50K
20908	77	Self-emp-not-inc	71676	Some-college	10	Widowed	Adm-clerical	Not-in-family	White	Female	0	1944	1	United-States	<=50K

3256 rows × 15 columns

python

df.sample(n = 5, random_state=10)

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	hours-per-week	native-country	income
11358	41	Private	134130	Bachelors	13	Married-civ-spouse	Sales	Husband	White	Male	0	50	United-States	>50K
10859	38	Private	107302	Some-college	10	Married-civ-spouse	Sales	Husband	White	Male	0	60	United-States	>50K
30948	24	Private	136687	HS-grad	9	Separated	Machine-op-inspct	Unmarried	Other	Female	0	40	United-States	<=50K
29811	35	Self-emp-inc	187053	Bachelors	13	Separated	Prof-specialty	Not-in-family	White	Female	0	50	United-States	<=50K
18408	69	?	254834	Bachelors	13	Married-civ-spouse	?	Husband	White	Male	10605	10	United-States	>50K

python

df.sample(n = 5, random_state=10)

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	hours-per-week	native-country	income
11358	41	Private	134130	Bachelors	13	Married-civ-spouse	Sales	Husband	White	Male	0	50	United-States	>50K
10859	38	Private	107302	Some-college	10	Married-civ-spouse	Sales	Husband	White	Male	0	60	United-States	>50K
30948	24	Private	136687	HS-grad	9	Separated	Machine-op-inspct	Unmarried	Other	Female	0	40	United-States	<=50K
29811	35	Self-emp-inc	187053	Bachelors	13	Separated	Prof-specialty	Not-in-family	White	Female	0	50	United-States	<=50K
18408	69	?	254834	Bachelors	13	Married-civ-spouse	?	Husband	White	Male	10605	10	United-States	>50K

df.index: índice

En el DataFrame df, no especificamos ningún índice especial, así que se creó uno por defecto a partir de una secuencia de números enteros. Puedes verlo de esta manera:

python

df.index

RangeIndex(start=0, stop=32560, step=1)

python

df.index.values

array([    0,     1,     2, ..., 32557, 32558, 32559], dtype=int64)

En efecto, el índice que pandas ha creado es un rango de números enteros consecutivos, desde 0 (incluido) hasta 23561 (sin incluir).

df.columns: nombres de las columnas

python

df.columns

Index(['age', 'workclass', 'fnlwgt', 'education', 'education-num',
       'marital-status', 'occupation', 'relationship', 'race', 'sex',
       'capital-gain', 'capital-loss', 'hours-per-week', 'native-country',
       'income'],
      dtype='object')

Selección de filas y columnas (indexación)

Casi todas las operaciones de procesamiento de datos implican seleccionar las columnas concretas sobre las que se desea trabajar, lo que se puede hacer de varias formas, ya sea con los métodos nativos del lenguaje Python o con los métodos de la librería pandas.

python

file_path = "C:/Users/user/Desktop/adult.data"

col_names = ["age", 
            "workclass",
            "fnlwgt", 
            "education",
            "education-num",
            "marital-status",
            "occupation",
            "relationship",
            "race",
            "sex",
            "capital-gain",
            "capital-loss",
            "hours-per-week",
            "native-country",
            "income"]


adult_data = pd.read_csv(filepath_or_buffer = file_path,
           header = 0,
           names = col_names)
adult_data

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	0	40	United-States	<=50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32555	27	Private	257302	Assoc-acdm	12	Married-civ-spouse	Tech-support	Wife	White	Female	0	0	38	United-States	<=50K
32556	40	Private	154374	HS-grad	9	Married-civ-spouse	Machine-op-inspct	Husband	White	Male	0	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	9	Widowed	Adm-clerical	Unmarried	White	Female	0	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	9	Married-civ-spouse	Exec-managerial	Wife	White	Female	15024	0	40	United-States	>50K

32560 rows × 15 columns

Métodos nativos del lenguaje Python

Selección de filas

python

adult_data[0:4]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	hours-per-week	native-country	income
0	50	Self-emp-not-inc	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	40	Cuba	<=50K

python

adult_data[0:-2]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
0	50	Self-emp-not-inc	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	0	40	United-States	<=50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32553	53	Private	321865	Masters	14	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	40	United-States	>50K
32554	22	Private	310152	Some-college	10	Never-married	Protective-serv	Not-in-family	White	Male	0	0	40	United-States	<=50K
32555	27	Private	257302	Assoc-acdm	12	Married-civ-spouse	Tech-support	Wife	White	Female	0	0	38	United-States	<=50K
32556	40	Private	154374	HS-grad	9	Married-civ-spouse	Machine-op-inspct	Husband	White	Male	0	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	9	Widowed	Adm-clerical	Unmarried	White	Female	0	0	40	United-States	<=50K

32558 rows × 15 columns

python

adult_data[:3]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	hours-per-week	native-country	income
0	50	Self-emp-not-inc	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	40	United-States	<=50K

Selección de columnas

El lenguaje Python, ya de por sí, proporciona algunas maneras de indexar datos, como por ejemplo accediendo a una propiedad de un objeto como si de un atributo se tratara. Si quisiéramos acceder a la columna occupation del DataFrame anterior, podríamos hacerlo de la siguiente manera:

python

adult_data.occupation

0           Exec-managerial
1         Handlers-cleaners
2         Handlers-cleaners
3            Prof-specialty
4           Exec-managerial
                ...        
32555          Tech-support
32556     Machine-op-inspct
32557          Adm-clerical
32558          Adm-clerical
32559       Exec-managerial
Name: occupation, Length: 32560, dtype: object

Por otro lado, de forma similar a como hacemos para acceder a los valores de un diccionario de Python, también podemos acceder a los valores de una columna usando el operador []:

python

adult_data["occupation"]

0           Exec-managerial
1         Handlers-cleaners
2         Handlers-cleaners
3            Prof-specialty
4           Exec-managerial
                ...        
32555          Tech-support
32556     Machine-op-inspct
32557          Adm-clerical
32558          Adm-clerical
32559       Exec-managerial
Name: occupation, Length: 32560, dtype: object

Por tanto, los dos métodos anteriores permiten seleccionar una Serie concreta de un DataFrame. Ambos son igual de válidos, pero el operador de indexación [] tiene la gran ventaja de que permite nombres de columnas con caracteres especiales. Es decir, si queremos acceder a la columna hours-per-week:

df.hours-per-week no funcionará (no sabe tratar los guiones, ni tampoco funcionaría si el nombre tuviera espacios en blanco).
df[“hours-per-week”] sí funcionará.

Comprobémoslo:

python

adult_data.hours-per-week

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
<ipython-input-29-3516cc3a2087> in <module>
----> 1 df.hours-per-week

~\anaconda3\envs\python-385\lib\site-packages\pandas\core\generic.py in __getattr__(self, name)
   5134             if self._info_axis._can_hold_identifiers_and_holds_name(name):
   5135                 return self[name]
-> 5136             return object.__getattribute__(self, name)
   5137 
   5138     def __setattr__(self, name: str, value) -> None:

AttributeError: 'DataFrame' object has no attribute 'hours'

python

adult_data["hours-per-week"]

0        13
1        40
2        40
3        40
4        40
         ..
32555    38
32556    40
32557    40
32558    20
32559    40
Name: hours-per-week, Length: 32560, dtype: int64

Si no solo queremos seleccionar una columna, sino un valor concreto dentro de esa columna, podemos acceder a él de la siguiente manera:

python

adult_data["hours-per-week"][32555]

Donde [1] indica la posición del valor dentro de la columna.

Métodos de pandas

Los métodos nativos de Python anteriores funcionan muy bien, pero la librería pandas tiene también sus propios métodos para seleccionar y acceder a columnas y valores concretos de un objeto.

python

adult_data

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	0	40	United-States	<=50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32555	27	Private	257302	Assoc-acdm	12	Married-civ-spouse	Tech-support	Wife	White	Female	0	0	38	United-States	<=50K
32556	40	Private	154374	HS-grad	9	Married-civ-spouse	Machine-op-inspct	Husband	White	Male	0	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	9	Widowed	Adm-clerical	Unmarried	White	Female	0	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	9	Married-civ-spouse	Exec-managerial	Wife	White	Female	15024	0	40	United-States	>50K

32560 rows × 15 columns

iloc: selección basada en posición

El primer método de pandas permite hacer selecciones basadas en un índice, es decir, en la posición numérica de un dato detro del conjunto de datos. Su sintaxis es data.iloc[filas,columnas], donde filas y columnas hacen referencia a la posición de las filas y columnas que se desea seleccionar.

Selección de filas

Tanto loc como iloc requieren indicar primero la fila y, en segundo lugar, la columna. Es decir, funcionan justo al revés que los métodos nativos de Python para indexación. Si no se especifican columnas concretas, se seleccionan todas las columnas de la fila. Por ejemplo, para seleccionar la primera fila del DataFrame con el que estamos trabajando:

python

adult_data.iloc[0]

age                                50
workclass                         NaN
fnlwgt                          83311
education                   Bachelors
education-num                      13
marital-status     Married-civ-spouse
occupation            Exec-managerial
relationship                  Husband
race                            White
sex                              Male
capital-gain                        0
capital-loss                        0
hours-per-week                     13
native-country          United-States
income                          <=50K
Name: 0, dtype: object

python

adult_data.iloc[32555]

age                                27
workclass                     Private
fnlwgt                         257302
education                  Assoc-acdm
education-num                      12
marital-status     Married-civ-spouse
occupation               Tech-support
relationship                     Wife
race                            White
sex                            Female
capital-gain                        0
capital-loss                        0
hours-per-week                     38
native-country          United-States
income                          <=50K
Name: 32555, dtype: object

python

adult_data[0:3]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	40	United-States	<=50K

python

adult_data.iloc[0:3]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	40	United-States	<=50K

Si se quiere seleccionar alguna fila empezando a contar desde el final del DataFrame, pueden usarse valores negativos:

python

adult_data.iloc[-1]

age                                52
workclass                Self-emp-inc
fnlwgt                         287927
education                     HS-grad
education-num                       9
marital-status     Married-civ-spouse
occupation            Exec-managerial
relationship                     Wife
race                            White
sex                            Female
capital-gain                    15024
capital-loss                        0
hours-per-week                     40
native-country          United-States
income                           >50K
Name: 32559, dtype: object

python

# Penúltima fila
adult_data.iloc[-2]

age                           22
workclass                Private
fnlwgt                    201490
education                HS-grad
education-num                  9
marital-status     Never-married
occupation          Adm-clerical
relationship           Own-child
race                       White
sex                         Male
capital-gain                   0
capital-loss                   0
hours-per-week                20
native-country     United-States
income                     <=50K
Name: 32558, dtype: object

Selección de columnas

Para seleccionar una columna completa con iloc, hacemos lo siguiente:

python

# Primera columna
adult_data.iloc[:, 0]

0        50
1        38
2        53
3        28
4        37
         ..
32555    27
32556    40
32557    58
32558    22
32559    52
Name: age, Length: 32560, dtype: int64

python

# Última columna
adult_data.iloc[:, -1]

0         <=50K
1         <=50K
2         <=50K
3         <=50K
4         <=50K
          ...  
32555     <=50K
32556      >50K
32557     <=50K
32558     <=50K
32559      >50K
Name: income, Length: 32560, dtype: object

Selección de filas y columnas

El operador “:”, por sí solo, ordena que se recojan “todos los valores”. Sin embargo, puede combinarse con otros selectores para referirse a un rango limitado de valores:

python

# Desde la primera fila (incluida) hasta la de índice 3 (sin incluir)
adult_data.iloc[0:3]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	40	United-States	<=50K

python

# Desde la primera fila (incluida) hasta la fila de índice 3 (sin incluir), pertenecientes a la primera columna
adult_data.iloc[0:3, 0]

0    50
1    38
2    53
Name: age, dtype: int64

python

# Todas las filas de las columnas con posiciones 0-3 (la 3 sin incluir)
adult_data.iloc[:, 0:3]

	age	workclass	fnlwgt
0	50	NaN	83311
1	38	Private	215646
2	53	Private	234721
3	28	Private	338409
4	37	Private	284582
...	...	...	...
32555	27	Private	257302
32556	40	Private	154374
32557	58	Private	151910
32558	22	Private	201490
32559	52	Self-emp-inc	287927

32560 rows × 3 columns

python

# Desde la fila 1 (incluida) hasta la 3 (sin incluir), pertenecientes a la primera columna
adult_data.iloc[1:3, 0]

1    38
2    53
Name: age, dtype: int64

python

# Primera, segunda y tercera fila
adult_data.iloc[[0, 1, 2]]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	40	United-States	<=50K

python

# Primera, segunda y tercera fila
adult_data.iloc[[0, 1, 2], :]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	40	United-States	<=50K

python

# Las filas 0, 1 y 2 de la primera columna
adult_data.iloc[[0,1,2], 0]

0    50
1    38
2    53
Name: age, dtype: int64

python

# Todas las filas de las columnas 0, 2 y 4
adult_data.iloc[:, [0, 2, 4]]

	age	fnlwgt	education-num
0	50	83311	13
1	38	215646	9
2	53	234721	7
3	28	338409	13
4	37	284582	14
...	...	...	...
32555	27	257302	12
32556	40	154374	9
32557	58	151910	9
32558	22	201490	9
32559	52	287927	9

32560 rows × 3 columns

También es posible utilizar números negativos en la selección, lo que implica que se empezará a contar desde el final de los valores:

python

# Fila 5 de un dataset, empezando a contar desde el final
adult_data.iloc[-5]

age                                27
workclass                     Private
fnlwgt                         257302
education                  Assoc-acdm
education-num                      12
marital-status     Married-civ-spouse
occupation               Tech-support
relationship                     Wife
race                            White
sex                            Female
capital-gain                        0
capital-loss                        0
hours-per-week                     38
native-country          United-States
income                          <=50K
Name: 32555, dtype: object

python

# Fila 5 de un dataset, empezando a contar desde el final
adult_data.iloc[-5, :]

age                                27
workclass                     Private
fnlwgt                         257302
education                  Assoc-acdm
education-num                      12
marital-status     Married-civ-spouse
occupation               Tech-support
relationship                     Wife
race                            White
sex                            Female
capital-gain                        0
capital-loss                        0
hours-per-week                     38
native-country          United-States
income                          <=50K
Name: 32555, dtype: object

loc: selección basada en etiquetas

python

df = pd.DataFrame({"Columna1": 10,
                  "Columna2": 20,
                  "Columna3": 30},
                 index = ["Fila1", "Fila2", "Fila3"])
df

	Columna1	Columna2	Columna3
Fila1	10	20	30
Fila2	10	20	30
Fila3	10	20	30

El segundo enfoque de pandas consiste en la selección basada en las etiquetas de los valores, en lugar de su posición.

python

# Primer valor de la columna "education-num"
df.loc["Fila1", "Columna1"]

python

# Todas las filas de las columnas "education", "marital-status" y "relationship"
adult_data.loc[:, ["education", "marital-status", "relationship"]]

	education	marital-status	relationship
0	Bachelors	Married-civ-spouse	Husband
1	HS-grad	Divorced	Not-in-family
2	11th	Married-civ-spouse	Husband
3	Bachelors	Married-civ-spouse	Wife
4	Masters	Married-civ-spouse	Wife
...	...	...	...
32555	Assoc-acdm	Married-civ-spouse	Wife
32556	HS-grad	Married-civ-spouse	Husband
32557	HS-grad	Widowed	Unmarried
32558	HS-grad	Never-married	Own-child
32559	HS-grad	Married-civ-spouse	Wife

32560 rows × 3 columns

python

# Primer valor de la columna "education"
adult_data.loc[0, "education"]

' Bachelors'

Como hemos visto, el método iloc icluye el primer elemento del rango pero excluye el último, es decir, 0:10 seleccionará los valores del 0 al 9, ambos incluidos. El método loc, sin embargo, incluye tanto el primer elemento del rango como el último. Por tanto, 0:10 seleccionará los valores del 0 al 10, ambos incluidos.

python

adult_data.loc[0:10, ["education", "marital-status", "relationship"]]

	education	marital-status	relationship
0	Bachelors	Married-civ-spouse	Husband
1	HS-grad	Divorced	Not-in-family
2	11th	Married-civ-spouse	Husband
3	Bachelors	Married-civ-spouse	Wife
4	Masters	Married-civ-spouse	Wife
5	9th	Married-spouse-absent	Not-in-family
6	HS-grad	Married-civ-spouse	Husband
7	Masters	Never-married	Not-in-family
8	Bachelors	Married-civ-spouse	Husband
9	Some-college	Married-civ-spouse	Husband
10	Bachelors	Married-civ-spouse	Husband

python

adult_data.iloc[0:10, :]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	40	United-States	<=50K
5	49	Private	160187	9th	5	Married-spouse-absent	Other-service	Not-in-family	Black	Female	0	16	Jamaica	<=50K
6	52	Self-emp-not-inc	209642	HS-grad	9	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	45	United-States	>50K
7	31	Private	45781	Masters	14	Never-married	Prof-specialty	Not-in-family	White	Female	14084	50	United-States	>50K
8	42	Private	159449	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	5178	40	United-States	>50K
9	37	Private	280464	Some-college	10	Married-civ-spouse	Exec-managerial	Husband	Black	Male	0	80	United-States	>50K

Selección basada en condiciones

El método loc de pandas no solo permite seleccionar filas y columnas por etiquetas, sino también en base a condiciones.

Imagina que estamos interesados en conocer los adultos del dataset que nunca se han casado y que, además trabajan 40 horas o más a la semana. En primer lugar, podríamos descubrir cuáles no se han casado nunca de la siguiente manera:

python

adult_data.head(20)

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	40	United-States	<=50K
5	49	Private	160187	9th	5	Married-spouse-absent	Other-service	Not-in-family	Black	Female	0	16	Jamaica	<=50K
6	52	Self-emp-not-inc	209642	HS-grad	9	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	45	United-States	>50K
7	31	Private	45781	Masters	14	Never-married	Prof-specialty	Not-in-family	White	Female	14084	50	United-States	>50K
8	42	Private	159449	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	5178	40	United-States	>50K
9	37	Private	280464	Some-college	10	Married-civ-spouse	Exec-managerial	Husband	Black	Male	0	80	United-States	>50K
10	30	State-gov	141297	Bachelors	13	Married-civ-spouse	Prof-specialty	Husband	Asian-Pac-Islander	Male	0	40	India	>50K
11	23	Private	122272	Bachelors	13	Never-married	Adm-clerical	Own-child	White	Female	0	30	United-States	<=50K
12	32	Private	205019	Assoc-acdm	12	Never-married	Sales	Not-in-family	Black	Male	0	50	United-States	<=50K
13	40	Private	121772	Assoc-voc	11	Married-civ-spouse	Craft-repair	Husband	Asian-Pac-Islander	Male	0	40	?	>50K
14	34	Private	245487	7th-8th	4	Married-civ-spouse	Transport-moving	Husband	Amer-Indian-Eskimo	Male	0	45	Mexico	<=50K
15	25	Self-emp-not-inc	176756	HS-grad	9	Never-married	Farming-fishing	Own-child	White	Male	0	35	United-States	<=50K
16	32	Private	186824	HS-grad	9	Never-married	Machine-op-inspct	Unmarried	White	Male	0	40	United-States	<=50K
17	38	Private	28887	11th	7	Married-civ-spouse	Sales	Husband	White	Male	0	50	United-States	<=50K
18	43	Self-emp-not-inc	292175	Masters	14	Divorced	Exec-managerial	Unmarried	White	Female	0	45	United-States	>50K
19	40	Private	193524	Doctorate	16	Married-civ-spouse	Prof-specialty	Husband	White	Male	0	60	United-States	>50K

python

adult_data.loc[:, "marital-status"] == "Never-married"

0        False
1        False
2        False
3        False
4        False
         ...  
32555    False
32556    False
32557    False
32558    False
32559    False
Name: marital-status, Length: 32560, dtype: bool

Esta operación produce una Serie con valores True/False en función de si la propiedad “marital-status” coincide con “Never-married” o no. Si introducimos esta operación en un loc, solo estaremos seleccionando los datos que cumplan la condición:

python

adult_data.loc[adult_data.loc[:, "marital-status"] == " Never-married"]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
7	31	Private	45781	Masters	14	Never-married	Prof-specialty	Not-in-family	White	Female	14084	0	50	United-States	>50K
11	23	Private	122272	Bachelors	13	Never-married	Adm-clerical	Own-child	White	Female	0	0	30	United-States	<=50K
12	32	Private	205019	Assoc-acdm	12	Never-married	Sales	Not-in-family	Black	Male	0	0	50	United-States	<=50K
15	25	Self-emp-not-inc	176756	HS-grad	9	Never-married	Farming-fishing	Own-child	White	Male	0	0	35	United-States	<=50K
16	32	Private	186824	HS-grad	9	Never-married	Machine-op-inspct	Unmarried	White	Male	0	0	40	United-States	<=50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32536	30	Private	345898	HS-grad	9	Never-married	Craft-repair	Not-in-family	Black	Male	0	0	46	United-States	<=50K
32547	65	Self-emp-not-inc	99359	Prof-school	15	Never-married	Prof-specialty	Not-in-family	White	Male	1086	0	60	United-States	<=50K
32552	32	Private	116138	Masters	14	Never-married	Tech-support	Not-in-family	Asian-Pac-Islander	Male	0	0	11	Taiwan	<=50K
32554	22	Private	310152	Some-college	10	Never-married	Protective-serv	Not-in-family	White	Male	0	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	0	20	United-States	<=50K

10682 rows × 15 columns

python

# La celda anterior devuelve un dataframe vacío, ¿por qué?
# Si aplicamos un unique(), vemos que ' Never-married' tiene un espacio en primera posición

python

adult_data.loc[:, "marital-status"].unique()

array([' Married-civ-spouse', ' Divorced', ' Married-spouse-absent',
       ' Never-married', ' Separated', ' Married-AF-spouse', ' Widowed'],
      dtype=object)

Hasta aquí, solo hemos seleccionado los adultos que no han estado casados, pero no aquellos que trabajan 40h o más a la semana.

python

adult_data.loc[(adult_data.loc[:, "marital-status"] == " Never-married") &
              (adult_data.loc[:, "hours-per-week"] >= 40)]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
7	31	Private	45781	Masters	14	Never-married	Prof-specialty	Not-in-family	White	Female	14084	0	50	United-States	>50K
12	32	Private	205019	Assoc-acdm	12	Never-married	Sales	Not-in-family	Black	Male	0	0	50	United-States	<=50K
16	32	Private	186824	HS-grad	9	Never-married	Machine-op-inspct	Unmarried	White	Male	0	0	40	United-States	<=50K
25	19	Private	168294	HS-grad	9	Never-married	Craft-repair	Own-child	White	Male	0	0	40	United-States	<=50K
29	23	Local-gov	190709	Assoc-acdm	12	Never-married	Protective-serv	Not-in-family	White	Male	0	0	52	United-States	<=50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32530	30	?	33811	Bachelors	13	Never-married	?	Not-in-family	Asian-Pac-Islander	Female	0	0	99	United-States	<=50K
32535	34	Private	160216	Bachelors	13	Never-married	Exec-managerial	Not-in-family	White	Female	0	0	55	United-States	>50K
32536	30	Private	345898	HS-grad	9	Never-married	Craft-repair	Not-in-family	Black	Male	0	0	46	United-States	<=50K
32547	65	Self-emp-not-inc	99359	Prof-school	15	Never-married	Prof-specialty	Not-in-family	White	Male	1086	0	60	United-States	<=50K
32554	22	Private	310152	Some-college	10	Never-married	Protective-serv	Not-in-family	White	Male	0	0	40	United-States	<=50K

6872 rows × 15 columns

Imagina ahora que queremos conocer los adultos que no han estado casados o que trabajan 40h o más a la semana.

python

adult_data.loc[(adult_data.loc[:, "marital-status"] == " Never-married") |
              (adult_data.loc[:, "hours-per-week"] >= 40)]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	0	40	United-States	<=50K
6	52	Self-emp-not-inc	209642	HS-grad	9	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	45	United-States	>50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32554	22	Private	310152	Some-college	10	Never-married	Protective-serv	Not-in-family	White	Male	0	0	40	United-States	<=50K
32556	40	Private	154374	HS-grad	9	Married-civ-spouse	Machine-op-inspct	Husband	White	Male	0	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	9	Widowed	Adm-clerical	Unmarried	White	Female	0	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	9	Married-civ-spouse	Exec-managerial	Wife	White	Female	15024	0	40	United-States	>50K

28607 rows × 15 columns

python

adult_data.loc[:, "marital-status"]

0         Married-civ-spouse
1                   Divorced
2         Married-civ-spouse
3         Married-civ-spouse
4         Married-civ-spouse
                ...         
32555     Married-civ-spouse
32556     Married-civ-spouse
32557                Widowed
32558          Never-married
32559     Married-civ-spouse
Name: marital-status, Length: 32560, dtype: object

python

adult_data["marital-status"]

0         Married-civ-spouse
1                   Divorced
2         Married-civ-spouse
3         Married-civ-spouse
4         Married-civ-spouse
                ...         
32555     Married-civ-spouse
32556     Married-civ-spouse
32557                Widowed
32558          Never-married
32559     Married-civ-spouse
Name: marital-status, Length: 32560, dtype: object

python

adult_data.loc[(adult_data.loc[:, "marital-status"] == " Never-married") |
              (adult_data.loc[:, "hours-per-week"] >= 40)]

python

adult_data.loc[(adult_data["marital-status"] == " Never-married") |
              (adult_data["hours-per-week"] >= 40)]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	0	40	United-States	<=50K
6	52	Self-emp-not-inc	209642	HS-grad	9	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	45	United-States	>50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32554	22	Private	310152	Some-college	10	Never-married	Protective-serv	Not-in-family	White	Male	0	0	40	United-States	<=50K
32556	40	Private	154374	HS-grad	9	Married-civ-spouse	Machine-op-inspct	Husband	White	Male	0	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	9	Widowed	Adm-clerical	Unmarried	White	Female	0	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	9	Married-civ-spouse	Exec-managerial	Wife	White	Female	15024	0	40	United-States	>50K

28607 rows × 15 columns

python

# Métodos nativos de Python
adult_data[(adult_data["marital-status"] == " Never-married") |
           (adult_data["hours-per-week"] >= 40)]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	0	40	United-States	<=50K
6	52	Self-emp-not-inc	209642	HS-grad	9	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	45	United-States	>50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32554	22	Private	310152	Some-college	10	Never-married	Protective-serv	Not-in-family	White	Male	0	0	40	United-States	<=50K
32556	40	Private	154374	HS-grad	9	Married-civ-spouse	Machine-op-inspct	Husband	White	Male	0	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	9	Widowed	Adm-clerical	Unmarried	White	Female	0	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	9	Married-civ-spouse	Exec-managerial	Wife	White	Female	15024	0	40	United-States	>50K

28607 rows × 15 columns

Pandas incluye también unos cuantos selectores específicos muy útiles, entre los cuales destaca isin. Este operador permite seleccionar los datos cuyos valores están dentro de una lista especificada. Por ejemplo, podríamos seleccionar los adultos que tiene nivel educativo “Bachelors” (como una licenciatura) o “Masters”:

python

value_list = [" Bachelors", " Masters"]
value_list

[' Bachelors', ' Masters']

python

# Antes de nada, vamos a asegurarnos de cómo están escritas ambas categorías en los datos

python

adult_data.loc[:, "education"].isin(value_list)

0         True
1        False
2        False
3         True
4         True
         ...  
32555    False
32556    False
32557    False
32558    False
32559    False
Name: education, Length: 32560, dtype: bool

python

adult_data.loc[adult_data.loc[:, "education"].isin(value_list)]

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	13	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	0	40	United-States	<=50K
7	31	Private	45781	Masters	14	Never-married	Prof-specialty	Not-in-family	White	Female	14084	0	50	United-States	>50K
8	42	Private	159449	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	5178	0	40	United-States	>50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32535	34	Private	160216	Bachelors	13	Never-married	Exec-managerial	Not-in-family	White	Female	0	0	55	United-States	>50K
32537	38	Private	139180	Bachelors	13	Divorced	Prof-specialty	Unmarried	Black	Female	15020	0	45	United-States	>50K
32543	31	Private	199655	Masters	14	Divorced	Other-service	Not-in-family	Other	Female	0	0	30	United-States	<=50K
32552	32	Private	116138	Masters	14	Never-married	Tech-support	Not-in-family	Asian-Pac-Islander	Male	0	0	11	Taiwan	<=50K
32553	53	Private	321865	Masters	14	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	40	United-States	>50K

7077 rows × 15 columns

python

adult_data.loc[:, "education"].unique()

array([' Bachelors', ' HS-grad', ' 11th', ' Masters', ' 9th',
       ' Some-college', ' Assoc-acdm', ' Assoc-voc', ' 7th-8th',
       ' Doctorate', ' Prof-school', ' 5th-6th', ' 10th', ' 1st-4th',
       ' Preschool', ' 12th'], dtype=object)

Asignación de datos

python

adult_data.head(3)

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	40	United-States	<=50K

Asignar datos a un DataFrame es muy sencillo, ya sea un valor constante o un conjunto de valores:

python

adult_data.loc[:, "education-num"] = "valor_constante"

python

adult_data.occupation = "occupation"

python

adult_data["marital-status"] = "estado"

python

adult_data["relationship"] = range(len(adult_data))

python

# adult_data["relationship"] = [10, 11, -5 , 3 ...]

python

adult_data["marital-status"] = adult_data["occupation"]

python

adult_data.loc[adult_data["education"] == " Bachelors", "marital-status"] = "X"

python

adult_data.loc[adult_data["education"] == " HS-grad", ["marital-status", "race"]] = "Y", "Z"

python

adult_data

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	valor_constante	X	occupation	0	White	Male	0	0	13	United-States	<=50K
1	38	Private	215646	HS-grad	valor_constante	Y	occupation	1	Z	Male	0	0	40	United-States	<=50K
2	53	Private	234721	11th	valor_constante	occupation	occupation	2	Black	Male	0	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	valor_constante	X	occupation	3	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	valor_constante	occupation	occupation	4	White	Female	0	0	40	United-States	<=50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32555	27	Private	257302	Assoc-acdm	valor_constante	occupation	occupation	32555	White	Female	0	0	38	United-States	<=50K
32556	40	Private	154374	HS-grad	valor_constante	Y	occupation	32556	Z	Male	0	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	valor_constante	Y	occupation	32557	Z	Female	0	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	valor_constante	Y	occupation	32558	Z	Male	0	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	valor_constante	Y	occupation	32559	Z	Female	15024	0	40	United-States	>50K

32560 rows × 15 columns

Como hemos visto, si la columna existe, se sobreescriben sus valores con el que le hayamos asignado. Si no existe, se crea una nueva columna:

python

adult_data["new_column"] = "New"

python

adult_data.loc[adult_data["sex"] == " Male", "nueva_columna"] = "M"

python

adult_data.loc[adult_data["sex"] != " Male", "nueva_columna2"] = "F"

python

adult_data["sex"].unique()

array([' Male', ' Female'], dtype=object)

python

adult_data

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income	new_column	nueva_columna	nueva_columna2
0	50	NaN	83311	Bachelors	valor_constante	X	occupation	0	White	Male	0	0	13	United-States	<=50K	New	M	NaN
1	38	Private	215646	HS-grad	valor_constante	Y	occupation	1	Z	Male	0	0	40	United-States	<=50K	New	M	NaN
2	53	Private	234721	11th	valor_constante	occupation	occupation	2	Black	Male	0	0	40	United-States	<=50K	New	M	NaN
3	28	Private	338409	Bachelors	valor_constante	X	occupation	3	Black	Female	0	0	40	Cuba	<=50K	New	NaN	F
4	37	Private	284582	Masters	valor_constante	occupation	occupation	4	White	Female	0	0	40	United-States	<=50K	New	NaN	F
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32555	27	Private	257302	Assoc-acdm	valor_constante	occupation	occupation	32555	White	Female	0	0	38	United-States	<=50K	New	NaN	F
32556	40	Private	154374	HS-grad	valor_constante	Y	occupation	32556	Z	Male	0	0	40	United-States	>50K	New	M	NaN
32557	58	Private	151910	HS-grad	valor_constante	Y	occupation	32557	Z	Female	0	0	40	United-States	<=50K	New	NaN	F
32558	22	Private	201490	HS-grad	valor_constante	Y	occupation	32558	Z	Male	0	0	20	United-States	<=50K	New	M	NaN
32559	52	Self-emp-inc	287927	HS-grad	valor_constante	Y	occupation	32559	Z	Female	15024	0	40	United-States	>50K	New	NaN	F

32560 rows × 18 columns

Manipulación del índice

Existe un método de pandas, set_index(), que permite establecer como índice el que nosotros queramos, como por ejemplo una columna:

python

adult_data

	age	workclass	fnlwgt	education	education-num	marital-status	occupation	relationship	race	sex	capital-gain	capital-loss	hours-per-week	native-country	income
0	50	NaN	83311	Bachelors	13	Married-civ-spouse	Exec-managerial	Husband	White	Male	0	0	13	United-States	<=50K
1	38	Private	215646	HS-grad	9	Divorced	Handlers-cleaners	Not-in-family	White	Male	0	0	40	United-States	<=50K
2	53	Private	234721	11th	7	Married-civ-spouse	Handlers-cleaners	Husband	Black	Male	0	0	40	United-States	<=50K
3	28	Private	338409	Bachelors	13	Married-civ-spouse	Prof-specialty	Wife	Black	Female	0	0	40	Cuba	<=50K
4	37	Private	284582	Masters	14	Married-civ-spouse	Exec-managerial	Wife	White	Female	0	0	40	United-States	<=50K
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
32555	27	Private	257302	Assoc-acdm	12	Married-civ-spouse	Tech-support	Wife	White	Female	0	0	38	United-States	<=50K
32556	40	Private	154374	HS-grad	9	Married-civ-spouse	Machine-op-inspct	Husband	White	Male	0	0	40	United-States	>50K
32557	58	Private	151910	HS-grad	9	Widowed	Adm-clerical	Unmarried	White	Female	0	0	40	United-States	<=50K
32558	22	Private	201490	HS-grad	9	Never-married	Adm-clerical	Own-child	White	Male	0	0	20	United-States	<=50K
32559	52	Self-emp-inc	287927	HS-grad	9	Married-civ-spouse	Exec-managerial	Wife	White	Female	15024	0	40	United-States	>50K

32560 rows × 15 columns

python

df = pd.DataFrame({"Columna1": [10, 20, 30],
                  "Columna2": [40, 50, 60],
                  "Columna3": 30},
                 index = ["Fila1", "Fila2", "Fila3"])
df

	Columna1	Columna2	Columna3
Fila1	10	40	30
Fila2	20	50	30
Fila3	30	60	30

python

nuevo_indice = ["Id1", "Id2", "Id3"]
nuevo_indice

['Id1', 'Id2', 'Id3']

python

df.set_index(nuevo_indice)

---------------------------------------------------------------------------
KeyError                                  Traceback (most recent call last)
<ipython-input-8-d4ffeb50a63e> in <module>
----> 1 df.set_index(nuevo_indice)

~\anaconda3\envs\python-385\lib\site-packages\pandas\core\frame.py in set_index(self, keys, drop, append, inplace, verify_integrity)
   4548 
   4549         if missing:
-> 4550             raise KeyError(f"None of {missing} are in the columns")
   4551 
   4552         if inplace:

KeyError: "None of ['Id1', 'Id2', 'Id3'] are in the columns"

python

nuevo_indice_arr = np.array(["Id1", "Id2", "Id3"])
nuevo_indice_arr

array(['Id1', 'Id2', 'Id3'], dtype='<U3')

python

df.set_index(nuevo_indice_arr)

	Columna1	Columna2	Columna3
Id1	10	40	30
Id2	20	50	30
Id3	30	60	30

python

df

	Columna1	Columna2	Columna3
Fila1	10	40	30
Fila2	20	50	30
Fila3	30	60	30

python

# Primera opción para conservar el resultado
df.set_index(nuevo_indice_arr, inplace=True)

python

df

	Columna1	Columna2	Columna3
Id1	10	40	30
Id2	20	50	30
Id3	30	60	30

python

# Primera opción para conservar el resultado
df = df.set_index(np.array(["i1", "i2", "i3"]), inplace=False)

python

df

	Columna1	Columna2	Columna3
i1	10	40	30
i2	20	50	30
i3	30	60	30

python

df.set_index("Columna3", inplace=True)

python

df

	Columna1	Columna2
Columna3
30	10	40
30	20	50
30	30	60

python

df["Columna4"] = [40, 40, 40]
df

	Columna1	Columna2	Columna4
Columna3
30	10	40	40
30	20	50	40
30	30	60	40

python

df.set_index("Columna4", inplace=True, verify_integrity=True)

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-20-93462f2e4f72> in <module>
----> 1 df.set_index("Columna4", inplace=True, verify_integrity=True)

~\anaconda3\envs\python-385\lib\site-packages\pandas\core\frame.py in set_index(self, keys, drop, append, inplace, verify_integrity)
   4600         if verify_integrity and not index.is_unique:
   4601             duplicates = index[index.duplicated()].unique()
-> 4602             raise ValueError(f"Index has duplicate keys: {duplicates}")
   4603 
   4604         # use set to handle duplicate column names gracefully in case of drop

ValueError: Index has duplicate keys: Int64Index([40], dtype='int64', name='Columna4')

python

df.index

Int64Index([30, 30, 30], dtype='int64', name='Columna3')

python

df.index.values

array([30, 30, 30], dtype=int64)

python

df.index.to_list()

[30, 30, 30]