DataFrame 数据处理字符转数值

from sklearn.neighbors import KNeighborsClassifier

from sklearn.model_selection import StratifiedKFold,KFold

sad26.png

workclass = df['workclass'].unique()
workclass

123.png

np.argwhere(workclass=='State-gov')[0][0]

>>>0

def conver(x):
    return np.argwhere(workclass==x)[0,0]

X['workclass'] = X['workclass'].map(conver)

234.png

X.columns

qwe17.png

clos = ['marital_status', 'occupation',
       'relationship', 'race', 'sex', 'native_country']

for clo in clos:
    u = X[clo].unique()
    
    def conver(x):
        return np.argwhere(u==x)[0,0]
    
    X[clo] = X[clo].map(conver)

12313.png

sKfold = StratifiedKFold(n_splits=20)

knn = KNeighborsClassifier(n_neighbors=5)
scores = []
for train,test in sKfold.split(X,y):
#     print(train.shape)
    knn.fit(X.iloc[train],y[train])
    
    s = knn.score(X.iloc[test],y[test])
    scores.append(s)

np.mean(scores)

>>>0.800651505347638