TIL 12일차 (22.01.17)

맹찬

|2022. 1. 18. 02:53

<안내>
필자도 배우는 입장이라 틀린점, 잘못된 점이 있을 수 있습니다.
그러니 지적, 피드백 환영합니다.

MISSION II¶

2. Kaggle에서 Dataset을 찾고, 이 Dataset에서 유의미한 Feature를 3개 이상 찾고 이를 시각화해봅시다.¶

In [1]:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

%matplotlib inline

In [4]:

egg_df = pd.read_csv('./GallusGallusDomesticus.csv')

In [6]:

egg_df

Out[6]:

	GallusID	GallusBreed	Day	Age	GallusWeight	GallusEggColor	GallusEggWeight	AmountOfFeed	EggsPerDay	GallusCombType	SunLightExposure	GallusClass	GallusLegShanksColor	GallusBeakColor	GallusEarLobesColor	GallusPlumage
0	Marans1	Marans	1	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
1	Marans1	Marans	2	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
2	Marans1	Marans	3	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
3	Marans1	Marans	4	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
4	Marans1	Marans	5	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
995	Ameraucana11	Ameraucana	1	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten
996	Ameraucana11	Ameraucana	2	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten
997	Ameraucana11	Ameraucana	3	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten
998	Ameraucana11	Ameraucana	4	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten
999	Ameraucana11	Ameraucana	5	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten

1000 rows × 16 columns

In [8]:

egg_df['GallusID'].unique()

Out[8]:

array(['Marans1', 'Marans2', 'Marans3', 'Marans4', 'Marans5', 'Marans6',
       'Marans7', 'Marans8', 'Marans9', 'Marans10', 'Marans11',
       'Marans12', 'Marans13', 'Marans14', 'Marans15', 'Marans16',
       'Marans17', 'Marans18', 'Marans19', 'Marans20', 'Marans21',
       'Marans22', 'Marans23', 'Marans24', 'Marans25', 'Marans26',
       'Marans27', 'Marans28', 'Marans29', 'Marans30', 'Marans31',
       'Marans32', 'Marans33', 'Marans34', 'Marans35', 'Marans36',
       'Marans37', 'Marans38', 'Marans39', 'Marans40', 'Marans41',
       'Marans42', 'Marans43', 'Marans44', 'Marans45', 'Marans46',
       'Marans47', 'Marans48', 'Marans49', 'Marans50', 'Marans51',
       'Marans52', 'Marans53', 'Marans54', 'Marans55', 'Marans56',
       'Marans57', 'Marans58', 'Marans59', 'Marans60', 'Marans61',
       'Marans62', 'Marans63', 'Marans64', 'Marans65', 'Marans66',
       'Marans67', 'Marans68', 'Marans69', 'Marans70', 'Marans71',
       'Marans72', 'Marans73', 'Marans74', 'Marans75', 'Marans76',
       'Marans77', 'Marans78', 'Marans79', 'Marans80', 'Marans81',
       'Marans82', 'Marans83', 'Marans84', 'Marans85', 'Marans86',
       'Marans87', 'Marans88', 'Marans89', 'Marans90', 'Marans91',
       'Marans92', 'Marans93', 'Marans94', 'Marans95', 'Marans96',
       'Marans97', 'Marans98', 'Marans99', 'Marans100', 'Marans101',
       'Marans102', 'Marans103', 'Marans104', 'Marans105', 'Marans106',
       'Marans107', 'Marans108', 'Marans109', 'Marans110', 'Marans111',
       'Marans112', 'Marans113', 'Marans114', 'Marans115', 'Marans116',
       'Marans117', 'Marans118', 'Marans119', 'Marans120', 'Marans121',
       'Marans122', 'Marans123', 'Marans124', 'Marans125', 'Marans126',
       'Marans127', 'Marans128', 'Marans129', 'Marans130', 'Marans131',
       'Marans132', 'Marans133', 'Marans134', 'Marans135', 'Marans136',
       'Marans137', 'Marans138', 'Marans139', 'Marans140', 'Marans141',
       'Marans142', 'Marans143', 'Marans144', 'Marans145', 'Marans146',
       'Marans147', 'Marans148', 'Marans149', 'Marans150', 'Marans151',
       'Marans152', 'Marans153', 'Marans154', 'Marans155', 'Marans156',
       'Marans157', 'Marans158', 'Marans159', 'Marans160', 'Marans161',
       'Marans162', 'Marans163', 'Marans164', 'Marans165', 'Marans166',
       'Marans167', 'Marans168', 'Marans169', 'Marans170', 'Marans171',
       'Marans172', 'Marans173', 'Marans174', 'Marans175', 'Marans176',
       'Marans177', 'Marans178', 'Marans179', 'Ameraucana1',
       'Ameraucana2', 'Ameraucana3', 'Ameraucana4', 'Ameraucana5',
       'Ameraucana6', 'Ameraucana7', 'Ameraucana8', 'Ameraucana9',
       'Ameraucana10', 'Ameraucana11'], dtype=object)

In [25]:

egg_df['Age'].unique()

Out[25]:

array([883, 684, 132, 226, 288, 125, 494, 984, 908, 619, 802, 725, 491,
       342, 134, 833, 474, 906, 442, 233, 118, 899, 866, 444, 425, 859,
       179, 469, 265, 567, 832, 460, 902, 124, 626, 825, 182, 635, 236,
       904, 648, 936, 979, 863, 809, 193, 896, 769, 913,  84, 660, 699,
       228, 102, 350, 407,  48, 990, 461, 664, 555, 929, 361, 761, 694,
       933, 400,  76, 332, 681, 344, 617,  91, 568, 187, 840, 224, 201,
       612, 594, 801, 447, 358, 692, 939, 732, 591, 589,  72, 393,  94,
       294, 466, 526, 615,  55, 395, 405, 717,  24, 559, 961,  95, 206,
       418, 186, 383, 200, 146, 262,  35, 147, 639, 266, 160, 707, 475,
       661, 753, 853,  71, 521,  61, 117, 907, 964, 421, 845, 814, 702,
       882,  92, 976, 529, 800, 250, 659, 867, 687, 488, 547, 576, 799,
       161, 854, 135, 893, 700, 143, 107, 554, 467, 705, 987, 556, 858,
       316, 795, 844, 514, 727, 208, 439, 204, 177, 901, 384], dtype=int64)

In [44]:

egg_df.describe()

Out[44]:

	Day	Age	GallusWeight	GallusEggWeight	AmountOfFeed	EggsPerDay	SunLightExposure
count	1000.00000	1000.000000	1000.000000	1000.000000	1000.000000	1000.000000	1000.000000
mean	3.25000	522.010000	2217.850000	43.427100	116.250000	0.965000	8.300000
std	1.78625	284.765045	438.544409	7.510839	7.514917	0.183872	1.269493
min	1.00000	24.000000	1500.000000	30.080000	100.000000	0.000000	5.000000
25%	2.00000	246.500000	1840.000000	36.895000	110.000000	1.000000	7.000000
50%	3.00000	527.500000	2170.000000	43.775000	116.000000	1.000000	8.000000
75%	4.00000	796.000000	2640.000000	50.020000	123.000000	1.000000	9.000000
max	10.00000	990.000000	3000.000000	58.930000	129.000000	1.000000	11.000000

In [45]:

egg_df.corr()

Out[45]:

	Day	Age	GallusWeight	GallusEggWeight	AmountOfFeed	EggsPerDay	SunLightExposure
Day	1.000000	0.001914	-0.082693	0.135436	-0.094145	0.026668	-0.077250
Age	0.001914	1.000000	0.052396	0.077701	0.128792	-0.304632	0.076941
GallusWeight	-0.082693	0.052396	1.000000	-0.101810	0.128234	0.095273	0.161182
GallusEggWeight	0.135436	0.077701	-0.101810	1.000000	-0.068021	-0.037039	-0.025482
AmountOfFeed	-0.094145	0.128792	0.128234	-0.068021	1.000000	-0.109570	0.007345
EggsPerDay	0.026668	-0.304632	0.095273	-0.037039	-0.109570	1.000000	0.045028
SunLightExposure	-0.077250	0.076941	0.161182	-0.025482	0.007345	0.045028	1.000000

In [35]:

marans_df = egg_df[egg_df['GallusBreed']=='Marans']
marans_df

Out[35]:

	GallusID	GallusBreed	Day	Age	GallusWeight	GallusEggColor	GallusEggWeight	AmountOfFeed	EggsPerDay	GallusCombType	SunLightExposure	GallusClass	GallusLegShanksColor	GallusBeakColor	GallusEarLobesColor	GallusPlumage
0	Marans1	Marans	1	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
1	Marans1	Marans	2	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
2	Marans1	Marans	3	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
3	Marans1	Marans	4	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
4	Marans1	Marans	5	883	3000	Brown	41.19	114	1	Single	7	Continental	White	White	NaN	Blue Copper
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
890	Marans179	Marans	1	844	2710	Brown	34.86	117	1	Single	8	Continental	White	White	NaN	Splash
891	Marans179	Marans	2	844	2710	Brown	34.86	117	1	Single	8	Continental	White	White	NaN	Splash
892	Marans179	Marans	3	844	2710	Brown	34.86	117	1	Single	8	Continental	White	White	NaN	Splash
893	Marans179	Marans	4	844	2710	Brown	34.86	117	1	Single	8	Continental	White	White	NaN	Splash
894	Marans179	Marans	5	844	2710	Brown	34.86	117	1	Single	8	Continental	White	White	NaN	Splash

895 rows × 16 columns

In [37]:

marans_df.describe()

Out[37]:

	Day	Age	GallusWeight	GallusEggWeight	AmountOfFeed	EggsPerDay	SunLightExposure
count	895.000000	895.000000	895.000000	895.000000	895.000000	895.000000	895.000000
mean	3.000000	522.346369	2249.106145	42.561788	116.849162	0.960894	8.385475
std	1.415004	287.309090	448.328096	7.291649	7.522869	0.193956	1.069200
min	1.000000	24.000000	1500.000000	30.080000	105.000000	0.000000	7.000000
25%	2.000000	250.000000	1880.000000	36.410000	110.000000	1.000000	7.000000
50%	3.000000	554.000000	2210.000000	42.890000	117.000000	1.000000	8.000000
75%	4.000000	800.000000	2670.000000	49.180000	123.000000	1.000000	9.000000
max	5.000000	990.000000	3000.000000	54.980000	129.000000	1.000000	10.000000

In [46]:

marans_df.corr()

Out[46]:

	Day	Age	GallusWeight	GallusEggWeight	AmountOfFeed	EggsPerDay	SunLightExposure
Day	1.000000e+00	3.909989e-17	-3.969028e-17	1.116959e-17	4.321181e-17	1.374448e-17	1.502128e-17
Age	3.909989e-17	1.000000e+00	4.492881e-02	8.251887e-02	1.647259e-01	-3.196183e-01	4.031080e-02
GallusWeight	-3.969028e-17	4.492881e-02	1.000000e+00	-3.101354e-02	7.529480e-02	1.127975e-01	1.298787e-01
GallusEggWeight	1.116959e-17	8.251887e-02	-3.101354e-02	1.000000e+00	4.097998e-02	-6.437112e-02	8.987063e-02
AmountOfFeed	4.321181e-17	1.647259e-01	7.529480e-02	4.097998e-02	1.000000e+00	-9.987379e-02	-1.779497e-02
EggsPerDay	1.374448e-17	-3.196183e-01	1.127975e-01	-6.437112e-02	-9.987379e-02	1.000000e+00	7.277204e-02
SunLightExposure	1.502128e-17	4.031080e-02	1.298787e-01	8.987063e-02	-1.779497e-02	7.277204e-02	1.000000e+00

In [47]:

ameraucana_df = egg_df[egg_df['GallusBreed']=='Ameraucana']
ameraucana_df

Out[47]:

	GallusID	GallusBreed	Day	Age	GallusWeight	GallusEggColor	GallusEggWeight	AmountOfFeed	EggsPerDay	GallusCombType	SunLightExposure	GallusClass	GallusLegShanksColor	GallusBeakColor	GallusEarLobesColor	GallusPlumage
895	Ameraucana1	Ameraucana	1	514	1740	Light Blue	47.12	114	1	Pea	6	NaN	NaN	White	White	Black
896	Ameraucana1	Ameraucana	2	514	1740	Light Blue	47.12	114	1	Pea	6	NaN	NaN	White	White	Black
897	Ameraucana1	Ameraucana	3	514	1740	Light Blue	47.12	114	1	Pea	6	NaN	NaN	White	White	Black
898	Ameraucana1	Ameraucana	4	514	1740	Light Blue	47.12	114	1	Pea	6	NaN	NaN	White	White	Black
899	Ameraucana1	Ameraucana	5	514	1740	Light Blue	47.12	114	1	Pea	6	NaN	NaN	White	White	Black
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
995	Ameraucana11	Ameraucana	1	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten
996	Ameraucana11	Ameraucana	2	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten
997	Ameraucana11	Ameraucana	3	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten
998	Ameraucana11	Ameraucana	4	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten
999	Ameraucana11	Ameraucana	5	384	1800	Bluish-Green	53.10	110	1	Pea	7	NaN	Slate Blue	Brown	White	Blue Wheaten

105 rows × 16 columns

In [40]:

ameraucana_df.describe()

Out[40]:

	Day	Age	GallusWeight	GallusEggWeight	AmountOfFeed	EggsPerDay	SunLightExposure
count	105.000000	105.000000	105.000000	105.000000	105.000000	105.0	105.000000
mean	5.380952	519.142857	1951.428571	50.802857	111.142857	1.0	7.571429
std	2.883577	263.345577	198.777168	4.845328	5.154215	0.0	2.248320
min	1.000000	177.000000	1720.000000	42.940000	100.000000	1.0	5.000000
25%	3.000000	208.000000	1780.000000	47.120000	109.000000	1.0	6.000000
50%	5.000000	469.000000	1880.000000	51.360000	112.000000	1.0	7.000000
75%	8.000000	727.000000	2100.000000	52.240000	116.000000	1.0	11.000000
max	10.000000	936.000000	2370.000000	58.930000	117.000000	1.0	11.000000

In [42]:

ameraucana_df.corr()

Out[42]:

	Day	Age	GallusWeight	GallusEggWeight	AmountOfFeed	EggsPerDay	SunLightExposure
Day	1.000000	0.021390	0.031753	-0.019761	0.009242	NaN	0.010594
Age	0.021390	1.000000	0.233892	0.105153	-0.316493	NaN	0.267494
GallusWeight	0.031753	0.233892	1.000000	-0.138104	0.333438	NaN	0.241276
GallusEggWeight	-0.019761	0.105153	-0.138104	1.000000	-0.548975	NaN	-0.210662
AmountOfFeed	0.009242	-0.316493	0.333438	-0.548975	1.000000	NaN	-0.197954
EggsPerDay	NaN	NaN	NaN	NaN	NaN	NaN	NaN
SunLightExposure	0.010594	0.267494	0.241276	-0.210662	-0.197954	NaN	1.000000

In [60]:

plt.scatter(x=ameraucana_df['GallusWeight'],y = ameraucana_df['GallusEggWeight'],c = 'Green')
plt.scatter(x=marans_df['GallusWeight'],y = marans_df['GallusEggWeight'],c = 'Red')
plt.show()

종에따른 닭의 무게와 달걀의 무게 관계¶

Ameraucana종은 Marans종에 비해 닭의 무게는 덜 나가고, 달걀의 무게는 더 많이 나감

In [91]:

egg_df[[
    'GallusEggColor','GallusClass','GallusLegShanksColor','GallusBeakColor','GallusEarLobesColor','GallusPlumage']]

Out[91]:

	GallusEggColor	GallusClass	GallusLegShanksColor	GallusBeakColor	GallusEarLobesColor	GallusPlumage
0	Brown	Continental	White	White	NaN	Blue Copper
1	Brown	Continental	White	White	NaN	Blue Copper
2	Brown	Continental	White	White	NaN	Blue Copper
3	Brown	Continental	White	White	NaN	Blue Copper
4	Brown	Continental	White	White	NaN	Blue Copper
...	...	...	...	...	...	...
995	Bluish-Green	NaN	Slate Blue	Brown	White	Blue Wheaten
996	Bluish-Green	NaN	Slate Blue	Brown	White	Blue Wheaten
997	Bluish-Green	NaN	Slate Blue	Brown	White	Blue Wheaten
998	Bluish-Green	NaN	Slate Blue	Brown	White	Blue Wheaten
999	Bluish-Green	NaN	Slate Blue	Brown	White	Blue Wheaten

1000 rows × 6 columns

In [68]:

egg_df.groupby(['GallusBreed','GallusLegShanksColor'])['GallusBreed'].count()

Out[68]:

GallusBreed  GallusLegShanksColor
Ameraucana   Slate Black              30
             Slate Blue               45
Marans       White                   895
Name: GallusBreed, dtype: int64

In [70]:

## Class 강( 종속과목강문계 에서)

egg_df.groupby(['GallusBreed','GallusClass'])['GallusBreed'].count()

Out[70]:

GallusBreed  GallusClass
Marans       Continental    895
Name: GallusBreed, dtype: int64

In [71]:

egg_df.groupby(['GallusBreed','GallusEarLobesColor'])['GallusBreed'].count()

Out[71]:

GallusBreed  GallusEarLobesColor
Ameraucana   Red                    50
             White                  35
Name: GallusBreed, dtype: int64

In [72]:

egg_df.groupby(['GallusBreed','GallusPlumage'])['GallusBreed'].count()

Out[72]:

GallusBreed  GallusPlumage
Ameraucana   Black             20
             Blue Wheaten      25
             Brown Red         20
             Buff              20
             Silver            10
             White             10
Marans       Blue             235
             Blue Copper      210
             Splash           230
             Splash Copper    220
Name: GallusBreed, dtype: int64

Breed 간의 차이 비교¶

In [92]:

## 닭의 종류와 EggColor

sns.catplot(x= 'GallusEggColor',col = 'GallusBreed' ,kind = 'count', data = egg_df)
plt.show()

In [93]:

## 닭의 종류와 Shanks 샹크스 아님 정강이 색깔임

sns.catplot(x= 'GallusLegShanksColor',col = 'GallusBreed' ,kind = 'count', data = egg_df)
plt.show()

In [94]:

## 닭의 종류와 beak Color 부리색

sns.catplot(x= 'GallusBeakColor',col = 'GallusBreed' ,kind = 'count', data = egg_df)
plt.show()

In [76]:

## 닭의 종류와 EarLobes 귓볼 색

sns.catplot(x= 'GallusEarLobesColor',col = 'GallusBreed' ,kind = 'count', data = egg_df)
plt.show()

In [83]:

## 닭의 종류와 Plumage 깃털층의 종류

sns.catplot(x= 'GallusPlumage',col = 'GallusBreed' ,kind = 'count', data = egg_df)
plt.show()

Breed 내에서 각 특징간 연관성 비교¶

In [99]:

## 정강이 색과 EggColor
## Maran 제외

sns.catplot(x= 'GallusEggColor',col = 'GallusLegShanksColor' ,kind = 'count', data = ameraucana_df)
plt.show()

In [98]:

## 부리색과 EggColor
## Maran 제외

sns.catplot(x= 'GallusEggColor',col = 'GallusBeakColor' ,kind = 'count', data = ameraucana_df)
plt.show()

In [81]:

## Ameraucana 종의 정강이 색과 깃털색 연관성
## Marans종은 정강이 색이 흰색으로 하나라서 제외

sns.catplot(x= 'GallusPlumage',col = 'GallusLegShanksColor' ,kind = 'count', data = ameraucana_df)
plt.show()

In [84]:

## Ameraucana 종의 귓볼색과 정강이 색의 연관성
## Marans종은 정강이 색이 흰색으로 하나라서 제외

sns.catplot(x= 'GallusEarLobesColor',col = 'GallusLegShanksColor' ,kind = 'count', data = ameraucana_df)
plt.show()

In [88]:

##  Ameraucana 종의 귓볼 색과 깃털색의 연관성
## Marans 종은 귓볼데이터가 없음
sns.catplot(x= 'GallusPlumage',col = 'GallusEarLobesColor' ,kind = 'count', data = egg_df)
plt.show()

In [89]:

egg_df[egg_df['GallusPlumage'] == 'Silver']

Out[89]:

	GallusID	GallusBreed	Day	Age	GallusWeight	GallusEggColor	GallusEggWeight	AmountOfFeed	EggsPerDay	GallusCombType	SunLightExposure	GallusClass	GallusLegShanksColor	GallusBeakColor	GallusEarLobesColor	GallusPlumage
935	Ameraucana5	Ameraucana	1	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver
936	Ameraucana5	Ameraucana	2	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver
937	Ameraucana5	Ameraucana	3	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver
938	Ameraucana5	Ameraucana	4	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver
939	Ameraucana5	Ameraucana	5	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver
940	Ameraucana5	Ameraucana	6	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver
941	Ameraucana5	Ameraucana	7	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver
942	Ameraucana5	Ameraucana	8	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver
943	Ameraucana5	Ameraucana	9	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver
944	Ameraucana5	Ameraucana	10	936	1880	Light Blue	51.36	100	1	Pea	11	NaN	NaN	NaN	NaN	Silver

종마다 특징이 있는 신체부위가 있음¶

Marans 종은 정강이가 모두 'White' 였고, 귓볼이 NA값이었고(없으니까 비운거겠지?), 깃털층은 'Blue'. 'Blue copper', 'Splash', 'Splash copper' 가 비슷한 비율로 존재 (그 중에선 'Blue'가 제일 많고, 'Blue Copper' 가 제일 적음)

Ameraucana 종은 정강이 색이 'Slate Black'과 'Slate Blue'를 가지고, 귓볼은 'White','Red'를 가짐
legshanks에 따라서 가지는 깃털이 살짝 다르고, 귓볼은 대체적으로 'Red'가 'White'보단 많았음.

특이하게도 Ameraucana종에서 깃털색은 'Silver'이면서 다른 값은 전부 Na값으로 표시된 종이 있음.

닭의 무게와 먹이의 양 상관관계¶

In [138]:

## 닭의 무게와 먹이의 양
plt.xlabel('GallusWeight')
plt.ylabel('AmountOfFeed')

plt.scatter(x='GallusWeight',y = 'AmountOfFeed',color = 'green',data = marans_df)
plt.scatter(x='GallusWeight',y = 'AmountOfFeed',color = 'red',data = ameraucana_df)
plt.show()

In [102]:

## 닭의 무게와 먹이의 양 
plt.scatter(x='GallusWeight',y = 'AmountOfFeed',color = 'green',data = marans_df)
plt.scatter(x='GallusWeight',y = 'AmountOfFeed',color = 'red',data = ameraucana_df)
plt.show()

닭의 무게와 먹는 양은 일정함을 알 수 있었음,¶

또한 Ameraucana종은 개체수가 충분하지않아, 확실하지는 않지만, 먹이를 적게먹고 더 무거운 달걀을 만들어내는 것으로 보임¶

닭의 나이와 / 닭의 무게, 달걀의 무게, 먹이의 양의 관계¶

In [140]:

## 닭의 나이와 닭의 무게
plt.scatter(x='Age',y = 'GallusWeight',color = 'green',data = marans_df)
plt.scatter(x='Age',y = 'GallusWeight',color = 'red',data = ameraucana_df)
plt.show()

In [141]:

## 닭의 나이와 달걀의 무게
plt.scatter(x='Age',y = 'GallusEggWeight',color = 'green',data = marans_df)
plt.scatter(x='Age',y = 'GallusEggWeight',color = 'red',data = ameraucana_df)
plt.show()

In [143]:

## 닭의 나이와 먹는 양 
plt.scatter(x='Age',y = 'AmountOfFeed',color = 'green',data = marans_df)
plt.scatter(x='Age',y = 'AmountOfFeed',color = 'red',data = ameraucana_df)
plt.show()

In [120]:

## 궁금한 점

sns.catplot (x = 'GallusWeight', y = 'AmountOfFeed',hue = 'GallusBreed', kind = 'point', data = egg_df)
plt.xticks(np.arange(1500,3000,100))
plt.show() 
# 하면 왜 xticks 가 잘 작동하지 않는거죠?

In [14]:

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns

%maplotlib inline

UsageError: Line magic function `%maplotlib` not found.

Core Mission¶

다음 질문에 답하시오.
- 캐릭터는 저마다 지능, 힘, ... 등 다양한 수치를 지니고 있다. 이러한 수치의 합이 가장 큰 캐릭터는 누구인가? 이를 보이기 위한 과정을 보여라.
- 좋은 캐릭터와 나쁜 캐릭터의 능력치들의 수치 분포를 알고 싶다. 이를 표현하기 위한 적절한 그래프를 선택해서 이를 위한 전처리를 진행하고, 시각화하여라.

Extra Mission¶

다음 질문에 답하시오.
- 캐릭터는 저마다 지능, 힘, ... 등 다양한 수치를 지니고 있다. 또한 각 캐릭터는 DC, 마블 등 다양한 코믹스 회사를 바탕으로 하고 있다. 어떤 코믹스 회사의 캐릭터들이 능력치 합의 평균이 가장 큰가? 이를 보이기 위한 과정을 보여라.
- 좋은 캐릭터와 나쁜 캐릭터가 격돌한다고 한다. 격돌하는 경우 캐릭터들의 능력치의 합의 평균이 큰 팀이 이긴다고 한다. 단, 불의를 못참는 중립 캐릭터들은 중립캐릭터가 없었을 당시에 열세인 팀에 가담한다. 이러한 상황일때 결과적으로 어떤 캐릭터 진영이 승리할 것인가? 이를 보이기 위한 과정을 보여라. 이 데이터를 이용해 진행하고 싶은 EDA 및 시각화가 있다면 자유롭게 진행하여라. 이 과제는 핵심 임무 이후에 진행되어야 한다.

MISSION I¶

수치의 합 가장 큰 영웅 찾기¶

접근법¶

DataFrame에서 Total 값이 큰 순서대로 배치한다면 수치합이 제일 큰 캐릭터를 알 수 있음

In [10]:

abengers_df = pd.read_csv("./charcters_stats.csv")
abengers_df

Out[10]:

	Name	Alignment	Intelligence	Strength	Speed	Durability	Power	Combat	Total
0	3-D Man	good	50	31	43	32	25	52	233
1	A-Bomb	good	38	100	17	80	17	64	316
2	Abe Sapien	good	88	14	35	42	35	85	299
3	Abin Sur	good	50	90	53	64	84	65	406
4	Abomination	bad	63	80	53	90	55	95	436
...	...	...	...	...	...	...	...	...	...
606	Yellowjacket	good	88	10	12	28	12	14	164
607	Yellowjacket II	good	50	10	35	28	31	28	182
608	Ymir	good	50	100	27	100	83	28	388
609	Zatanna	good	75	10	23	28	100	56	292
610	Zoom	bad	50	10	100	28	72	28	288

611 rows × 9 columns

In [13]:

abengers_df.sort_values(by= ['Total'],ascending = False)

Out[13]:

	Name	Alignment	Intelligence	Strength	Speed	Durability	Power	Combat	Total
361	Martian Manhunter	good	100	100	96	100	100	85	581
242	General Zod	bad	94	100	96	100	94	95	579
535	Superboy-Prime	bad	94	100	100	100	100	85	579
537	Superman	good	100	100	100	100	94	85	579
16	Amazo	bad	75	100	100	100	100	100	575
...	...	...	...	...	...	...	...	...	...
462	Renata Soliz	good	1	1	1	1	0	1	5
137	Captain Mar-vell	good	1	1	1	1	0	1	5
136	Captain Epic	good	1	1	1	1	0	1	5
466	Ripcord	good	1	1	1	1	0	1	5
335	Lady Deathstrike	bad	1	1	1	1	0	1	5

611 rows × 9 columns

수치분포¶

접근¶

Alignment가 good인 그룹과 bad인 그룹을 나누고
능력치의 index? 종류를 x축에, 그 능력치의 양을 y축으로 넣는 그래프를 만들자
(scatter plot이나 catplot정도?)

In [16]:

good_df = abengers_df[abengers_df['Alignment'] == 'good']
good_df

Out[16]:

	Name	Alignment	Intelligence	Strength	Speed	Durability	Power	Combat	Total
0	3-D Man	good	50	31	43	32	25	52	233
1	A-Bomb	good	38	100	17	80	17	64	316
2	Abe Sapien	good	88	14	35	42	35	85	299
3	Abin Sur	good	50	90	53	64	84	65	406
6	Adam Monroe	good	63	10	12	100	71	64	320
...	...	...	...	...	...	...	...	...	...
604	X-Man	good	88	53	53	95	92	84	465
606	Yellowjacket	good	88	10	12	28	12	14	164
607	Yellowjacket II	good	50	10	35	28	31	28	182
608	Ymir	good	50	100	27	100	83	28	388
609	Zatanna	good	75	10	23	28	100	56	292

432 rows × 9 columns

In [18]:

bad_df = abengers_df[abengers_df['Alignment'] == 'bad']
bad_df

Out[18]:

	Name	Alignment	Intelligence	Strength	Speed	Durability	Power	Combat	Total
4	Abomination	bad	63	80	53	90	55	95	436
5	Abraxas	bad	88	100	83	99	100	56	526
11	Air-Walker	bad	50	85	100	85	100	40	460
16	Amazo	bad	75	100	100	100	100	100	575
17	Ammo	bad	1	1	1	1	0	1	5
...	...	...	...	...	...	...	...	...	...
586	Warp	bad	38	10	23	28	63	50	212
590	Weapon XI	bad	1	1	1	1	0	1	5
593	Willis Stryker	bad	38	16	23	28	41	60	206
605	Yellow Claw	bad	1	1	1	1	0	1	5
610	Zoom	bad	50	10	100	28	72	28	288

165 rows × 9 columns

In [24]:

sns.catplot(bad_df)

c:\python39\lib\site-packages\seaborn\_decorators.py:36: FutureWarning: Pass the following variable as a keyword arg: x. From version 0.12, the only valid positional argument will be `data`, and passing other arguments without an explicit keyword will result in an error or misinterpretation.
  warnings.warn(

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
~\AppData\Local\Temp/ipykernel_12736/2703098325.py in <module>
----> 1 sns.catplot(bad_df)

c:\python39\lib\site-packages\seaborn\_decorators.py in inner_f(*args, **kwargs)
     44             )
     45         kwargs.update({k: arg for k, arg in zip(sig.parameters, args)})
---> 46         return f(**kwargs)
     47     return inner_f
     48 

c:\python39\lib\site-packages\seaborn\categorical.py in catplot(x, y, hue, data, row, col, col_wrap, estimator, ci, n_boot, units, seed, order, hue_order, row_order, col_order, kind, height, aspect, orient, color, palette, legend, legend_out, sharex, sharey, margin_titles, facet_kws, **kwargs)
   3790     p = _CategoricalPlotter()
   3791     p.require_numeric = plotter_class.require_numeric
-> 3792     p.establish_variables(x_, y_, hue, data, orient, order, hue_order)
   3793     if (
   3794         order is not None

c:\python39\lib\site-packages\seaborn\categorical.py in establish_variables(self, x, y, hue, data, orient, order, hue_order, units)
    154 
    155             # Figure out the plotting orientation
--> 156             orient = infer_orient(
    157                 x, y, orient, require_numeric=self.require_numeric
    158             )

c:\python39\lib\site-packages\seaborn\_core.py in infer_orient(x, y, orient, require_numeric)
   1309     """
   1310 
-> 1311     x_type = None if x is None else variable_type(x)
   1312     y_type = None if y is None else variable_type(y)
   1313 

c:\python39\lib\site-packages\seaborn\_core.py in variable_type(vector, boolean_type)
   1227 
   1228     # Special-case all-na data, which is always "numeric"
-> 1229     if pd.isna(vector).all():
   1230         return "numeric"
   1231 

c:\python39\lib\site-packages\pandas\core\generic.py in __nonzero__(self)
   1535     @final
   1536     def __nonzero__(self):
-> 1537         raise ValueError(
   1538             f"The truth value of a {type(self).__name__} is ambiguous. "
   1539             "Use a.empty, a.bool(), a.item(), a.any() or a.all()."

ValueError: The truth value of a Series is ambiguous. Use a.empty, a.bool(), a.item(), a.any() or a.all().

내일 할 것 : 어벤저스 df 분석 마무리 3주차 수업 듣기

어벤져스 df에서 각각의 능력치이름를 모아서 x축에 넣고, 그 능력치의 값을 y축에 모을 생각을 해보자.

## 궁금한 점

sns.catplot (x = 'GallusWeight', y = 'AmountOfFeed',hue = 'GallusBreed', kind = 'point', data = egg_df)
plt.xticks(np.arange(1500,3000,100))
plt.show() 
# 하면 왜 xticks 가 잘 작동하지 않는거죠?

'TIL > [겨울방학 부트캠프]TIL' 카테고리의 다른 글

TIL 14일차 (22.01.19) (0)	2022.01.20
TIL 13일 (22.01.18) (0)	2022.01.18
TIL 11일차 (22.01.16) (0)	2022.01.17
TIL 10일차 (22.01.14) (0)	2022.01.15
TIL 9일차 (22.01.13) (0)	2022.01.14