Support Vector Machine - bin - diabates dataset

서포트벡터머신 - 이진분류

• 피마인디언 당뇨병 데이터셋

 

In [3]:

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

import warnings
warnings.filterwarnings('ignore')

1. 데이터 준비

 

In [4]:

from google.colab import drive
drive.mount('/content/drive')

 

 

Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount("/content/drive", force_remount=True).

 

In [5]:

df = pd.read_csv("/content/drive/MyDrive/SKT FLY AI/2주차/diabetes.csv")

 

In [6]:

df.head()

 

Out[6]:

	Pregnancies	Glucose	BloodPressure	SkinThickness	Insulin	BMI	DiabetesPedigreeFunction	Age	Outcome
0	6	148	72	35	0	33.6	0.627	50	1
1	1	85	66	29	0	26.6	0.351	31	0
2	8	183	64	0	0	23.3	0.672	32	1
3	1	89	66	23	94	28.1	0.167	21	0
4	0	137	40	35	168	43.1	2.288	33	1

 

 

In [7]:

df.info()

 

 

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 768 entries, 0 to 767
Data columns (total 9 columns):
 #   Column                    Non-Null Count  Dtype  
---  ------                    --------------  -----  
 0   Pregnancies               768 non-null    int64  
 1   Glucose                   768 non-null    int64  
 2   BloodPressure             768 non-null    int64  
 3   SkinThickness             768 non-null    int64  
 4   Insulin                   768 non-null    int64  
 5   BMI                       768 non-null    float64
 6   DiabetesPedigreeFunction  768 non-null    float64
 7   Age                       768 non-null    int64  
 8   Outcome                   768 non-null    int64  
dtypes: float64(2), int64(7)
memory usage: 54.1 KB

 

In [8]:

df.describe().T

 

Out[8]:

	count	mean	std	min	25%	50%	75%	max
Pregnancies	768.0	3.845052	3.369578	0.000	1.00000	3.0000	6.00000	17.00
Glucose	768.0	120.894531	31.972618	0.000	99.00000	117.0000	140.25000	199.00
BloodPressure	768.0	69.105469	19.355807	0.000	62.00000	72.0000	80.00000	122.00
SkinThickness	768.0	20.536458	15.952218	0.000	0.00000	23.0000	32.00000	99.00
Insulin	768.0	79.799479	115.244002	0.000	0.00000	30.5000	127.25000	846.00
BMI	768.0	31.992578	7.884160	0.000	27.30000	32.0000	36.60000	67.10
DiabetesPedigreeFunction	768.0	0.471876	0.331329	0.078	0.24375	0.3725	0.62625	2.42
Age	768.0	33.240885	11.760232	21.000	24.00000	29.0000	41.00000	81.00
Outcome	768.0	0.348958	0.476951	0.000	0.00000	0.0000	1.00000	1.00

 

 

 

In [9]:

df.Outcome.value_counts()

 

Out[9]:

0    500
1    268
Name: Outcome, dtype: int64

 

 

2. 테스트데이터 분리¶

 

In [10]:

y = df.Outcome
y

 

Out[10]:

0      1
1      0
2      1
3      0
4      1
      ..
763    0
764    0
765    0
766    1
767    0
Name: Outcome, Length: 768, dtype: int64

 

In [11]:

X = df.drop('Outcome',axis=1)
X

 

Out[11]:

	Pregnancies	Glucose	BloodPressure	SkinThickness	Insulin	BMI	DiabetesPedigreeFunction	Age
0	6	148	72	35	0	33.6	0.627	50
1	1	85	66	29	0	26.6	0.351	31
2	8	183	64	0	0	23.3	0.672	32
3	1	89	66	23	94	28.1	0.167	21
4	0	137	40	35	168	43.1	2.288	33
...	...	...	...	...	...	...	...	...
763	10	101	76	48	180	32.9	0.171	63
764	2	122	70	27	0	36.8	0.340	27
765	5	121	72	23	112	26.2	0.245	30
766	1	126	60	0	0	30.1	0.349	47
767	1	93	70	31	0	30.4	0.315	23

768 rows × 8 columns

 

 

 

 

In [12]:

from sklearn.model_selection import train_test_split

X_train, X_test, y_train, y_test = train_test_split(X,y,test_size = 0.2, random_state = 42,stratify=y)

 

In [13]:

print(X_train.shape, X_test.shape, y_train.shape, y_test.shape)

 

 

(614, 8) (154, 8) (614,) (154,)

 

In [14]:

y_train.value_counts()

 

Out[14]:

0    400
1    214
Name: Outcome, dtype: int64

 

In [15]:

y_test.value_counts()

 

Out[15]:

0    100
1     54
Name: Outcome, dtype: int64

3. 전처리

 

결측치 체크

 

In [16]:

X_train.isna().sum()

 

Out[16]:

Pregnancies                 0
Glucose                     0
BloodPressure               0
SkinThickness               0
Insulin                     0
BMI                         0
DiabetesPedigreeFunction    0
Age                         0
dtype: int64

 

이상치 체크

• 0인값을 가진 샘플을 지우기

 

In [17]:

df1 = X_train.copy()
df1.shape

 

Out[17]:

(614, 8)

 

In [18]:

df1.columns

 

Out[18]:

Index(['Pregnancies', 'Glucose', 'BloodPressure', 'SkinThickness', 'Insulin',
       'BMI', 'DiabetesPedigreeFunction', 'Age'],
      dtype='object')

 

In [19]:

cols = ['Glucose', 'BloodPressure', 'SkinThickness', 'Insulin',
       'BMI']

 

In [20]:

df1 = df1[cols].replace(0,np.nan)   # 0을 결측치로 대체

df1.isna().sum()

 

Out[20]:

Glucose            4
BloodPressure     23
SkinThickness    175
Insulin          290
BMI                9
dtype: int64

 

In [21]:

df1 = df1.dropna()
df1.shape

# 결과값을 봤을 때 데이터의 유실이 너무 많음 -> 다른 방안으로 대체

 

Out[21]:

(322, 5)

 

 

• 0인 값을 적당한 값으로 채워넣는 방법 -> median 값으로 대체

 

In [22]:

df2 = X_train.copy()
df2.shape

 

Out[22]:

(614, 8)

 

In [23]:

df2[cols] = df2[cols].replace(0,np.nan)   # 0을 결측치로 대체

df2.isna().sum()

 

Out[23]:

Pregnancies                   0
Glucose                       4
BloodPressure                23
SkinThickness               175
Insulin                     290
BMI                           9
DiabetesPedigreeFunction      0
Age                           0
dtype: int64

 

In [24]:

df2['Glucose'] = df2['Glucose'].fillna(df2['Glucose'].median())
df2['BloodPressure'] = df2['BloodPressure'].fillna(df2['BloodPressure'].median())
df2['SkinThickness'] = df2['SkinThickness'].fillna(df2['SkinThickness'].median())
df2['Insulin'] = df2['Insulin'].fillna(df2['Insulin'].median())
df2['BMI'] = df2['BMI'].fillna(df2['BMI'].median())

 

In [25]:

G_m = df['Glucose'].median()
B_m = df['BloodPressure'].median()
S_m = df['SkinThickness'].median()
I_m = df['Insulin'].median()
Bm_m = df['BMI'].median()

 

In [26]:

df2.describe()

 

Out[26]:

	Pregnancies	Glucose	BloodPressure	SkinThickness	Insulin	BMI	DiabetesPedigreeFunction	Age
count	614.000000	614.000000	614.000000	614.000000	614.000000	614.000000	614.000000	614.000000
mean	3.819218	121.671010	72.140065	29.042345	137.705212	32.448208	0.477428	33.366450
std	3.314148	30.003794	12.275119	8.891855	78.764767	6.824122	0.330300	11.833438
min	0.000000	56.000000	24.000000	7.000000	15.000000	18.200000	0.084000	21.000000
25%	1.000000	99.000000	64.000000	25.000000	120.000000	27.625000	0.245000	24.000000
50%	3.000000	117.000000	72.000000	29.000000	125.000000	32.400000	0.382500	29.000000
75%	6.000000	140.000000	80.000000	32.000000	130.000000	36.500000	0.639250	41.000000
max	17.000000	199.000000	122.000000	99.000000	744.000000	67.100000	2.329000	81.000000

 

 

 

In [27]:

df2[cols] = df2[cols].fillna(df[cols])

 

In [28]:

df2.describe()

 

Out[28]:

	Pregnancies	Glucose	BloodPressure	SkinThickness	Insulin	BMI	DiabetesPedigreeFunction	Age
count	614.000000	614.000000	614.000000	614.000000	614.000000	614.000000	614.000000	614.000000
mean	3.819218	121.671010	72.140065	29.042345	137.705212	32.448208	0.477428	33.366450
std	3.314148	30.003794	12.275119	8.891855	78.764767	6.824122	0.330300	11.833438
min	0.000000	56.000000	24.000000	7.000000	15.000000	18.200000	0.084000	21.000000
25%	1.000000	99.000000	64.000000	25.000000	120.000000	27.625000	0.245000	24.000000
50%	3.000000	117.000000	72.000000	29.000000	125.000000	32.400000	0.382500	29.000000
75%	6.000000	140.000000	80.000000	32.000000	130.000000	36.500000	0.639250	41.000000
max	17.000000	199.000000	122.000000	99.000000	744.000000	67.100000	2.329000	81.000000

 

 

In [29]:

X_train = df2.copy()

 

 

스케일링

 

In [30]:

from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()
X_train_s = scaler.fit_transform(X_train)

X_train_s

 

Out[30]:

array([[-0.85135507, -1.05642747, -0.82674004, ..., -0.76969431,
         0.31079384, -0.79216928],
       [ 0.35657564,  0.14439907,  0.47777235, ..., -0.41771394,
        -0.11643851,  0.56103382],
       [-0.5493724 , -0.55608308, -1.15286813, ...,  0.35957603,
        -0.76486207, -0.70759409],
       ...,
       [-0.85135507, -0.82293342, -0.17448384, ...,  0.82888319,
        -0.78607218, -0.28471812],
       [ 1.86648903, -0.35594533, -0.17448384, ..., -0.72569676,
        -1.01938346,  0.56103382],
       [ 0.05459296,  0.74481233, -1.15286813, ..., -0.43237979,
        -0.57700104,  0.30730824]])

 

In [31]:

scaler.mean_

 

Out[31]:

array([  3.81921824, 121.67100977,  72.14006515,  29.04234528,
       137.70521173,  32.44820847,   0.47742834,  33.36644951])

 

In [32]:

y_train = y_train.values
type(y_train)

 

Out[32]:

numpy.ndarray

 

 

4. 학습

• SVC 베이스 모델

 

In [33]:

from sklearn.svm import SVC

clf = SVC(random_state = 42)
clf = clf.fit(X_train_s,y_train)

 

 

5. 예측

테스트 데이터 전처리 먼저

• 0인 값을 위에서 계산한 값으로 넣는 작업
• 스케일링

 

In [34]:

X_test.head(5)

 

Out[34]:

	Pregnancies	Glucose	BloodPressure	SkinThickness	Insulin	BMI	DiabetesPedigreeFunction	Age
44	7	159	64	0	0	27.4	0.294	40
672	10	68	106	23	49	35.5	0.285	47
700	2	122	76	27	200	35.9	0.483	26
630	7	114	64	0	0	27.4	0.732	34
81	2	74	0	0	0	0.0	0.102	22

 

 

In [35]:

X_test = X_test.replace(0,np.nan)
X_test

 

Out[35]:

	Pregnancies	Glucose	BloodPressure	SkinThickness	Insulin	BMI	DiabetesPedigreeFunction	Age
44	7.0	159.0	64.0	NaN	NaN	27.4	0.294	40
672	10.0	68.0	106.0	23.0	49.0	35.5	0.285	47
700	2.0	122.0	76.0	27.0	200.0	35.9	0.483	26
630	7.0	114.0	64.0	NaN	NaN	27.4	0.732	34
81	2.0	74.0	NaN	NaN	NaN	NaN	0.102	22
...	...	...	...	...	...	...	...	...
32	3.0	88.0	58.0	11.0	54.0	24.8	0.267	22
637	2.0	94.0	76.0	18.0	66.0	31.6	0.649	23
593	2.0	82.0	52.0	22.0	115.0	28.5	1.699	25
425	4.0	184.0	78.0	39.0	277.0	37.0	0.264	31
273	1.0	71.0	78.0	50.0	45.0	33.2	0.422	21

154 rows × 8 columns

 

 

In [36]:

X_test['Glucose'] = X_test['Glucose'].fillna(G_m)
X_test['BloodPressure'] = X_test['BloodPressure'].fillna(B_m)
X_test['SkinThickness'] = X_test['SkinThickness'].fillna(S_m)
X_test['Insulin'] = X_test['Insulin'].fillna(I_m)
X_test['BMI'] = X_test['BMI'].fillna(Bm_m )

 

In [37]:

X_test['Pregnancies'] = X_test['Pregnancies'].fillna(X_test['Pregnancies'].median())

 

In [38]:

X_test

 

Out[38]:

	Pregnancies	Glucose	BloodPressure	SkinThickness	Insulin	BMI	DiabetesPedigreeFunction	Age
44	7.0	159.0	64.0	23.0	30.5	27.4	0.294	40
672	10.0	68.0	106.0	23.0	49.0	35.5	0.285	47
700	2.0	122.0	76.0	27.0	200.0	35.9	0.483	26
630	7.0	114.0	64.0	23.0	30.5	27.4	0.732	34
81	2.0	74.0	72.0	23.0	30.5	32.0	0.102	22
...	...	...	...	...	...	...	...	...
32	3.0	88.0	58.0	11.0	54.0	24.8	0.267	22
637	2.0	94.0	76.0	18.0	66.0	31.6	0.649	23
593	2.0	82.0	52.0	22.0	115.0	28.5	1.699	25
425	4.0	184.0	78.0	39.0	277.0	37.0	0.264	31
273	1.0	71.0	78.0	50.0	45.0	33.2	0.422	21

154 rows × 8 columns

 

 

In [39]:

X_test.isna().sum()

 

Out[39]:

Pregnancies                 0
Glucose                     0
BloodPressure               0
SkinThickness               0
Insulin                     0
BMI                         0
DiabetesPedigreeFunction    0
Age                         0
dtype: int64

 

In [40]:

from sklearn.preprocessing import StandardScaler

scaler = StandardScaler()
X_test_s = scaler.fit_transform(X_test)
X_test_s

 

Out[40]:

array([[ 0.76197764,  1.16484536, -0.8288433 , ..., -0.71876183,
        -0.46562514,  0.63400979],
       [ 1.72381827, -1.66920722,  2.88630463, ...,  0.4261539 ,
        -0.49253293,  1.24533587],
       [-0.84109008,  0.01253827,  0.23262754, ...,  0.48269295,
         0.09943846, -0.58864236],
       ...,
       [-0.84109008, -1.23319913, -1.89031414, ..., -0.56327945,
         3.73497988, -0.67597466],
       [-0.19986299,  1.94343123,  0.40953934, ...,  0.63817533,
        -0.55531777, -0.15198088],
       [-1.16170362, -1.57577692,  0.40953934, ...,  0.10105437,
        -0.08293657, -1.02530385]])

 

In [41]:

y_pred = clf.predict(X_test_s)

 

In [42]:

from sklearn.metrics import accuracy_score, recall_score, precision_score,f1_score


acc = accuracy_score(y_test,y_pred) # 정답값, 예측

print(f'optimal_accuracy: {acc}')

 

 

optimal_accuracy: 0.7402597402597403

 

 

모델 튜닝

C, gamma,kernel,degrees 하이퍼 파라미터 튜닝

 

In [43]:

from sklearn.model_selection import GridSearchCV

params = {
    "C" : [0.1 ,1 ],
    'gamma' : [0.01,0.1],
    'degree' : [2,3],
    'kernel' : ['linear','poly']
}

grid_cv = GridSearchCV(SVC(),params,cv=5,refit= True, verbose=3)
grid_cv.fit(X_train_s,y_train)

 

 

Fitting 5 folds for each of 16 candidates, totalling 80 fits
[CV 1/5] END C=0.1, degree=2, gamma=0.01, kernel=linear;, score=0.764 total time=   0.0s
[CV 2/5] END C=0.1, degree=2, gamma=0.01, kernel=linear;, score=0.756 total time=   0.0s
[CV 3/5] END C=0.1, degree=2, gamma=0.01, kernel=linear;, score=0.756 total time=   0.0s
[CV 4/5] END C=0.1, degree=2, gamma=0.01, kernel=linear;, score=0.780 total time=   0.0s
[CV 5/5] END C=0.1, degree=2, gamma=0.01, kernel=linear;, score=0.811 total time=   0.0s
[CV 1/5] END C=0.1, degree=2, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 2/5] END C=0.1, degree=2, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 3/5] END C=0.1, degree=2, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 4/5] END C=0.1, degree=2, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 5/5] END C=0.1, degree=2, gamma=0.01, kernel=poly;, score=0.656 total time=   0.0s
[CV 1/5] END C=0.1, degree=2, gamma=0.1, kernel=linear;, score=0.764 total time=   0.0s
[CV 2/5] END C=0.1, degree=2, gamma=0.1, kernel=linear;, score=0.756 total time=   0.0s
[CV 3/5] END C=0.1, degree=2, gamma=0.1, kernel=linear;, score=0.756 total time=   0.0s
[CV 4/5] END C=0.1, degree=2, gamma=0.1, kernel=linear;, score=0.780 total time=   0.0s
[CV 5/5] END C=0.1, degree=2, gamma=0.1, kernel=linear;, score=0.811 total time=   0.0s
[CV 1/5] END C=0.1, degree=2, gamma=0.1, kernel=poly;, score=0.634 total time=   0.0s
[CV 2/5] END C=0.1, degree=2, gamma=0.1, kernel=poly;, score=0.659 total time=   0.0s
[CV 3/5] END C=0.1, degree=2, gamma=0.1, kernel=poly;, score=0.650 total time=   0.0s
[CV 4/5] END C=0.1, degree=2, gamma=0.1, kernel=poly;, score=0.650 total time=   0.0s
[CV 5/5] END C=0.1, degree=2, gamma=0.1, kernel=poly;, score=0.656 total time=   0.0s
[CV 1/5] END C=0.1, degree=3, gamma=0.01, kernel=linear;, score=0.764 total time=   0.0s
[CV 2/5] END C=0.1, degree=3, gamma=0.01, kernel=linear;, score=0.756 total time=   0.0s
[CV 3/5] END C=0.1, degree=3, gamma=0.01, kernel=linear;, score=0.756 total time=   0.0s
[CV 4/5] END C=0.1, degree=3, gamma=0.01, kernel=linear;, score=0.780 total time=   0.0s
[CV 5/5] END C=0.1, degree=3, gamma=0.01, kernel=linear;, score=0.811 total time=   0.0s
[CV 1/5] END C=0.1, degree=3, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 2/5] END C=0.1, degree=3, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 3/5] END C=0.1, degree=3, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 4/5] END C=0.1, degree=3, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 5/5] END C=0.1, degree=3, gamma=0.01, kernel=poly;, score=0.656 total time=   0.0s
[CV 1/5] END C=0.1, degree=3, gamma=0.1, kernel=linear;, score=0.764 total time=   0.0s
[CV 2/5] END C=0.1, degree=3, gamma=0.1, kernel=linear;, score=0.756 total time=   0.0s
[CV 3/5] END C=0.1, degree=3, gamma=0.1, kernel=linear;, score=0.756 total time=   0.0s
[CV 4/5] END C=0.1, degree=3, gamma=0.1, kernel=linear;, score=0.780 total time=   0.0s
[CV 5/5] END C=0.1, degree=3, gamma=0.1, kernel=linear;, score=0.811 total time=   0.0s
[CV 1/5] END C=0.1, degree=3, gamma=0.1, kernel=poly;, score=0.659 total time=   0.0s
[CV 2/5] END C=0.1, degree=3, gamma=0.1, kernel=poly;, score=0.659 total time=   0.0s
[CV 3/5] END C=0.1, degree=3, gamma=0.1, kernel=poly;, score=0.699 total time=   0.0s
[CV 4/5] END C=0.1, degree=3, gamma=0.1, kernel=poly;, score=0.691 total time=   0.0s
[CV 5/5] END C=0.1, degree=3, gamma=0.1, kernel=poly;, score=0.730 total time=   0.0s
[CV 1/5] END C=1, degree=2, gamma=0.01, kernel=linear;, score=0.772 total time=   0.0s
[CV 2/5] END C=1, degree=2, gamma=0.01, kernel=linear;, score=0.748 total time=   0.0s
[CV 3/5] END C=1, degree=2, gamma=0.01, kernel=linear;, score=0.756 total time=   0.0s
[CV 4/5] END C=1, degree=2, gamma=0.01, kernel=linear;, score=0.772 total time=   0.0s
[CV 5/5] END C=1, degree=2, gamma=0.01, kernel=linear;, score=0.803 total time=   0.0s
[CV 1/5] END C=1, degree=2, gamma=0.01, kernel=poly;, score=0.642 total time=   0.0s
[CV 2/5] END C=1, degree=2, gamma=0.01, kernel=poly;, score=0.650 total time=   0.1s
[CV 3/5] END C=1, degree=2, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 4/5] END C=1, degree=2, gamma=0.01, kernel=poly;, score=0.650 total time=   0.1s
[CV 5/5] END C=1, degree=2, gamma=0.01, kernel=poly;, score=0.656 total time=   0.0s
[CV 1/5] END C=1, degree=2, gamma=0.1, kernel=linear;, score=0.772 total time=   0.0s
[CV 2/5] END C=1, degree=2, gamma=0.1, kernel=linear;, score=0.748 total time=   0.0s
[CV 3/5] END C=1, degree=2, gamma=0.1, kernel=linear;, score=0.756 total time=   0.0s
[CV 4/5] END C=1, degree=2, gamma=0.1, kernel=linear;, score=0.772 total time=   0.0s
[CV 5/5] END C=1, degree=2, gamma=0.1, kernel=linear;, score=0.803 total time=   0.0s
[CV 1/5] END C=1, degree=2, gamma=0.1, kernel=poly;, score=0.659 total time=   0.0s
[CV 2/5] END C=1, degree=2, gamma=0.1, kernel=poly;, score=0.715 total time=   0.0s
[CV 3/5] END C=1, degree=2, gamma=0.1, kernel=poly;, score=0.683 total time=   0.0s
[CV 4/5] END C=1, degree=2, gamma=0.1, kernel=poly;, score=0.715 total time=   0.0s
[CV 5/5] END C=1, degree=2, gamma=0.1, kernel=poly;, score=0.656 total time=   0.0s
[CV 1/5] END C=1, degree=3, gamma=0.01, kernel=linear;, score=0.772 total time=   0.1s
[CV 2/5] END C=1, degree=3, gamma=0.01, kernel=linear;, score=0.748 total time=   0.0s
[CV 3/5] END C=1, degree=3, gamma=0.01, kernel=linear;, score=0.756 total time=   0.0s
[CV 4/5] END C=1, degree=3, gamma=0.01, kernel=linear;, score=0.772 total time=   0.0s
[CV 5/5] END C=1, degree=3, gamma=0.01, kernel=linear;, score=0.803 total time=   0.0s
[CV 1/5] END C=1, degree=3, gamma=0.01, kernel=poly;, score=0.642 total time=   0.0s
[CV 2/5] END C=1, degree=3, gamma=0.01, kernel=poly;, score=0.650 total time=   0.1s
[CV 3/5] END C=1, degree=3, gamma=0.01, kernel=poly;, score=0.650 total time=   0.0s
[CV 4/5] END C=1, degree=3, gamma=0.01, kernel=poly;, score=0.650 total time=   0.1s
[CV 5/5] END C=1, degree=3, gamma=0.01, kernel=poly;, score=0.656 total time=   0.1s
[CV 1/5] END C=1, degree=3, gamma=0.1, kernel=linear;, score=0.772 total time=   0.0s
[CV 2/5] END C=1, degree=3, gamma=0.1, kernel=linear;, score=0.748 total time=   0.0s
[CV 3/5] END C=1, degree=3, gamma=0.1, kernel=linear;, score=0.756 total time=   0.0s
[CV 4/5] END C=1, degree=3, gamma=0.1, kernel=linear;, score=0.772 total time=   0.0s
[CV 5/5] END C=1, degree=3, gamma=0.1, kernel=linear;, score=0.803 total time=   0.0s
[CV 1/5] END C=1, degree=3, gamma=0.1, kernel=poly;, score=0.707 total time=   0.0s
[CV 2/5] END C=1, degree=3, gamma=0.1, kernel=poly;, score=0.707 total time=   0.0s
[CV 3/5] END C=1, degree=3, gamma=0.1, kernel=poly;, score=0.691 total time=   0.0s
[CV 4/5] END C=1, degree=3, gamma=0.1, kernel=poly;, score=0.724 total time=   0.0s
[CV 5/5] END C=1, degree=3, gamma=0.1, kernel=poly;, score=0.779 total time=   0.0s

Out[43]:

GridSearchCV(cv=5, estimator=SVC(),
             param_grid={'C': [0.1, 1], 'degree': [2, 3], 'gamma': [0.01, 0.1],
                         'kernel': ['linear', 'poly']},
             verbose=3)

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

 

In [45]:

grid_cv.best_estimator_

 

Out[45]:

SVC(C=0.1, degree=2, gamma=0.01, kernel='linear')

In a Jupyter environment, please rerun this cell to show the HTML representation or trust the notebook.
On GitHub, the HTML representation is unable to render, please try loading this page with nbviewer.org.

 

In [46]:

grid_cv.best_score_   # 최고의 모델 정확도

 

Out[46]:

0.7736771957883513

 

In [47]:

grid_cv.best_params_

 

Out[47]:

{'C': 0.1, 'degree': 2, 'gamma': 0.01, 'kernel': 'linear'}

베스트 모델 학습

 

In [50]:

clf = SVC(C= 0.1, degree=2, gamma = 0.01, kernel = 'linear')
clf = clf.fit(X_train_s,y_train)

 

모델 저장

 

In [52]:

import pickle


with open('svc_c_1_degree_2_gamma_001_rbf.pickle','wb') as f:
  pickle.dump(clf,f)

 

In [54]:

# 스케일러

with open('stanard_scaler.pickle','wb') as f:
  pickle.dump(scaler,f)

 

 

저장된 모델 사용하기

 

In [ ]:

with open('/content/svc_c_1_degree_2_gamma_001_rbf.pickle','rb') as f:
  c = pickle.read(f)