Python Pandas DataFrame, 상관관계, 상관계수

예제

df = pd.read_csv('Salary_Data.csv')

YearsExperience	Salary
0	1.1	39343.0
1	1.3	46205.0
2	1.5	37731.0
3	2.0	43525.0
4	2.2	39891.0
5	2.9	56642.0
6	3.0	60150.0
7	3.2	54445.0
8	3.2	64445.0
9	3.7	57189.0
10	3.9	63218.0
11	4.0	55794.0
12	4.0	56957.0
13	4.1	57081.0
14	4.5	61111.0
15	4.9	67938.0
16	5.1	66029.0
17	5.3	83088.0
18	5.9	81363.0
19	6.0	93940.0
20	6.8	91738.0
21	7.1	98273.0
22	7.9	101302.0
23	8.2	113812.0
24	8.7	109431.0
25	9.0	105582.0
26	9.5	116969.0
27	9.6	112635.0
28	10.3	122391.0
29	10.5	121872.0

df.corr()

YearsExperience	Salary
YearsExperience	1.000000	0.978242
Salary	0.978242	1.000000

예제

df = pd.read_csv('customer_data.csv')

label	id	fea_1	fea_2	fea_3	fea_4	fea_5	fea_6	fea_7	fea_8	fea_9	fea_10	fea_11
0	1	54982665	5	1245.5	3	77000.0	2	15	5	109	5	151300	244.948974
1	0	59004779	4	1277.0	1	113000.0	2	8	-1	100	3	341759	207.173840
2	0	58990862	7	1298.0	1	110000.0	2	11	-1	101	5	72001	1.000000
3	1	58995168	7	1335.5	1	151000.0	2	11	5	110	3	60084	1.000000
4	0	54987320	7	NaN	2	59000.0	2	11	5	108	4	450081	197.403141
...	...	...	...	...	...	...	...	...	...	...	...	...	...
1120	0	58988196	5	1289.0	1	173000.0	2	15	5	112	3	350702	200.000000
1121	0	58987926	5	NaN	2	50000.0	2	15	5	108	4	450000	169.000000
1122	0	58995381	7	1220.0	3	76000.0	2	11	2	90	5	71002	1.000000
1123	0	58998054	4	1250.0	3	137000.0	2	8	5	90	5	72000	1.000000
1124	0	54989781	4	1415.0	3	93000.0	2	8	5	113	4	151300	273.861279

c = df.corr()

label	id	fea_1	fea_2	fea_3	fea_4	fea_5	fea_6	fea_7	fea_8	fea_9	fea_10	fea_11
label	1.000000e+00	-0.024619	0.052071	-0.078808	0.055658	-0.131459	0.017293	0.028907	-0.012271	-0.021251	5.590642e-17	0.000881	0.002444
id	-2.461878e-02	1.000000	-0.094888	0.037017	-0.011086	-0.015092	-0.046856	-0.066836	-0.027143	0.025433	-1.304576e-02	0.023954	0.029668
fea_1	5.207118e-02	-0.094888	1.000000	-0.012046	0.004391	-0.105110	-0.020977	0.346629	-0.030144	0.049951	-5.500809e-02	0.108505	0.095053
fea_2	-7.880778e-02	0.037017	-0.012046	1.000000	-0.246883	0.284455	0.003515	0.001176	0.003682	-0.022591	9.718013e-02	-0.029273	0.059464
fea_3	5.565787e-02	-0.011086	0.004391	-0.246883	1.000000	-0.119909	-0.017062	0.161899	0.253059	0.025249	1.285708e-01	-0.117266	0.066774
fea_4	-1.314593e-01	-0.015092	-0.105110	0.284455	-0.119909	1.000000	0.035862	-0.137654	-0.046294	-0.105451	1.000482e-01	0.066553	0.039042
fea_5	1.729286e-02	-0.046856	-0.020977	0.003515	-0.017062	0.035862	1.000000	0.055280	-0.017898	0.003813	-2.165627e-02	0.105127	-0.101328
fea_6	2.890698e-02	-0.066836	0.346629	0.001176	0.161899	-0.137654	0.055280	1.000000	-0.010971	0.028963	6.183006e-02	0.216362	0.128467
fea_7	-1.227143e-02	-0.027143	-0.030144	0.003682	0.253059	-0.046294	-0.017898	-0.010971	1.000000	0.078722	-1.652961e-02	-0.245453	0.021797
fea_8	-2.125122e-02	0.025433	0.049951	-0.022591	0.025249	-0.105451	0.003813	0.028963	0.078722	1.000000	-1.009981e-01	0.095880	0.133031
fea_9	5.590642e-17	-0.013046	-0.055008	0.097180	0.128571	0.100048	-0.021656	0.061830	-0.016530	-0.100998	1.000000e+00	-0.038176	-0.048769
fea_10	8.808212e-04	0.023954	0.108505	-0.029273	-0.117266	0.066553	0.105127	0.216362	-0.245453	0.095880	-3.817559e-02	1.000000	0.271108
fea_11	2.444255e-03	0.029668	0.095053	0.059464	0.066774	0.039042	-0.101328	0.128467	0.021797	0.133031	-4.876862e-02	0.271108	1.000000

uc = c.unstack()

uc.sort_values(kind='quicksort', ascending=False) # quicksort 방식 정렬, 내림차순

sorted = uc.sort_values(kind='quicksort', ascending=False) # quicksort 방식 정렬, 내림차순

sorted.head(20)

label   label     1.000000
id      id        1.000000
fea_10  fea_10    1.000000
fea_9   fea_9     1.000000
fea_8   fea_8     1.000000
fea_7   fea_7     1.000000
fea_6   fea_6     1.000000
fea_4   fea_4     1.000000
fea_3   fea_3     1.000000
fea_2   fea_2     1.000000
fea_1   fea_1     1.000000
fea_5   fea_5     1.000000
fea_11  fea_11    1.000000
fea_6   fea_1     0.346629
fea_1   fea_6     0.346629
fea_2   fea_4     0.284455
fea_4   fea_2     0.284455
fea_11  fea_10    0.271108
fea_10  fea_11    0.271108
fea_7   fea_3     0.253059
dtype: float64

예제

plt.scatter(df['fea_6'],df['fea_1'])

사인파형 곡선 만들기

# 사인파형 곡선이 표시되도록 라인차트를 그려보세요

# x : 0 ~ 360

# y : np.sin()

deg = np.arange(0, 360)

rad = np.radians(deg)

sinv = np.sin(rad)

cosv = np.cos(rad)

plt.scatter(deg, sinv, c='b', s=2, label='Sine Value')

plt.scatter(deg, cosv, c='r', s=2, label='Cosine Value')

plt.xlabel('Degree')

plt.ylabel('Trig.values')

plt.legend()

plt.show()

arr = np.random.normal(5,1,100) # 평균 5, 표준편차 1, 갯수 100

plt.hist(arr) # 히스토그램(데이터 빈도수 표시)

plt.show()