[Visualization] Basic_for _visualization

 

 

 

In [6]:

import numpy as np
import pandas as pd

 

 

numerical

 

In [11]:

num_df = pd.DataFrame(np.random.randint(0, 100, size=(10,5)), columns = list('ABCDE'))
num_df

 

Out[11]:

	A	B	C	D	E
0	5	9	66	69	87
1	18	87	31	59	42
2	30	69	31	13	38
3	90	96	19	59	91
4	34	95	27	56	69
5	28	36	26	34	53
6	57	55	96	6	16
7	48	30	19	8	58
8	6	72	65	76	61
9	96	72	0	77	93

 

 

categorical

 

In [12]:

cat_df = pd.DataFrame({"Color": ['Red','Blue','Green','Yellow'], 'Brand':['nike','samsung','naver','kakao']})
cat_df

 

Out[12]:

	Color	Brand
0	Red	nike
1	Blue	samsung
2	Green	naver
3	Yellow	kakao

 

 

Missing values

 

In [15]:

nan_data = {'example': [np.nan, np.nan, 10 ,20, 30, 50, 100,np.nan,np.nan,1, 2, 3, 4, 5, np.nan, np.nan, np.nan]}
nan_df = pd.DataFrame(nan_data,columns=['example'])

 

In [18]:

nan_df.isnull().values.any()

 

Out[18]:

True

 

In [19]:

nan_df['example'].isnull().values.any()

 

Out[19]:

True

 

In [31]:

clean_df = nan_df.fillna(0); clean_df.head(5)

 

Out[31]:

	example
0	0.0
1	0.0
2	10.0
3	20.0
4	30.0

 

In [25]:

nan_df.fillna(method='ffill').head(5)

 

Out[25]:

	example
0	NaN
1	NaN
2	10.0
3	20.0
4	30.0

 

In [30]:

nan_df.fillna(method='bfill').head(5)

 

Out[30]:

	example
0	10.0
1	10.0
2	10.0
3	20.0
4	30.0

 

In [27]:

df_drop = nan_df.dropna()
df_drop.head()

 

Out[27]:

	example
2	10.0
3	20.0
4	30.0
5	50.0
6	100.0

 

In [28]:

df_drop.describe()

 

Out[28]:

	example
count	10.000000
mean	22.500000
std	31.433351
min	1.000000
25%	3.250000
50%	7.500000
75%	27.500000
max	100.000000

 

 

Encoding

 

In [32]:

pd.get_dummies(cat_df)

 

Out[32]:

	Color_Blue	Color_Green	Color_Red	Color_Yellow	Brand_kakao	Brand_naver	Brand_nike	Brand_samsung
0	0	0	1	0	0	0	1	0
1	1	0	0	0	0	0	0	1
2	0	1	0	0	0	1	0	0
3	0	0	0	1	1	0	0	0

 

 

Converting Data Types

 

In [33]:

clean_df.dtypes

 

Out[33]:

example    float64
dtype: object

 

In [34]:

df_string = clean_df.astype('string')
df_string.dtypes

 

Out[34]:

example    string
dtype: object

 

In [35]:

df_int = clean_df.astype('int')
df_int.dtypes

 

Out[35]:

example    int64
dtype: object