Python 将字符串转换为唯一的数值或Numpy
我是一名Python初学者，正在深入学习。这对你们大多数人来说可能很容易，但我怎么能做到呢 如何将以下对象转换为唯一的数值Python 将字符串转换为唯一的数值或Numpy,python,pandas,tensorflow,deep-learning,keras,Python,Pandas,Tensorflow,Deep Learning,Keras,我是一名Python初学者，正在深入学习。这对你们大多数人来说可能很容易，但我怎么能做到呢 如何将以下对象转换为唯一的数值 df['city'].unique() array(['LIMA', 'VACAVILLE', 'CINCINNATI', 'GLASGOW', 'BOWLING GREEN', 'LANCASTER', 'HOUSTON', 'SPRINGFIELD', 'RAPID CITY', 'FORT WORTH', 'LAREDO', 'NEW YORK', 'C
df['city'].unique()
array(['LIMA', 'VACAVILLE', 'CINCINNATI', 'GLASGOW', 'BOWLING GREEN',
'LANCASTER', 'HOUSTON', 'SPRINGFIELD', 'RAPID CITY', 'FORT WORTH',
'LAREDO', 'NEW YORK', 'CHARLESTON', 'PITTSBURGH',
'WEST VALLEY CITY', 'CAYCE', 'HOT SPRINGS NATIO', 'CANTON',
'FORT WAYNE', 'DU BOIS', 'DAYTON', 'MASON CITY', 'WASHINGTON',
'LAKE OSWEGO', 'FAYETTEVILLE', 'SALT LAKE CITY', 'KNOXVILLE',
'TURLOCK', 'MCALLEN', 'CENTERVILLE', 'ROCHESTER', 'OKLAHOMA CITY',
'GAUTIER', 'DOYLESTOWN', 'ATLANTA', 'MEADVILLE', 'FORT MYERS',
'ERIE', 'BEAUMONT', 'JACKSON', 'CLARKSVILLE', 'BETHLEHEM',
'SAN ANTONIO', 'LAS VEGAS', 'ATHENS', 'SAN LUIS OBISPO', 'SEATTLE',
'BRADENTON', 'TINLEY PARK', 'HUNTLEY', 'SYRACUSE', 'WHEELWRIGHT',
'TOWSON', 'YONKERS', 'ARDEN HILLS', 'MARION', 'LIVONIA',
'COLORADO SPRINGS', 'CURWENSVILLE', 'SAINT CHARLES', 'PETERSBURG',
'SCOTTSDALE', 'SILVER SPRING', 'PORTLAND', 'BIRMINGHAM',
'CEDARVILLE', 'CLERMONT', 'ASHEVILLE', 'SHREVEPORT', 'DRAPER',
'WAVERLY', 'CANANDAIGUA', 'MOUNT PLEASANT', 'MARIETTA', 'MANKATO',
'HARLINGEN', 'HATCH', 'MOBILE', 'POULSBO', 'GARDEN GROVE',
'GIG HARBOR', 'OCONOMOWOC', 'MOUNT MORRIS', 'ORLANDO',
'DODGE CITY', 'DILLSBURG', 'HUNTSVILLE', 'KANSAS CITY',
'JACKSONVILLE', 'DULUTH', 'CITRUS HEIGHTS', 'ONEONTA', 'LOS LUNAS',
'GIBSONIA', 'ROBINSON', 'VERNON HILLS', 'PHOENIX', 'DESTIN',
'SHEPHERD', 'BROOKLYN', 'PLANO', 'WINTERS', 'JAMAICA', 'POWAY',
'LEXINGTON', 'UPLAND', 'NEW ALBANY', 'GREENVILLE',
'JEFFERSON CITY', 'ARLINGTON', 'BUFFALO', 'LOS ANGELES',
'CHARLOTTE', 'WEST LAFAYETTE', 'GARY', 'COOPERSTOWN', 'GREAT BEND',
'DAVISON', 'SMYRNA', 'MISSOURI CITY', 'MEMPHIS',
'FORT WALTON BEACH', 'KISSIMMEE', 'BATAVIA', 'OLDSMAR', 'WYNNE',
'ASHVILLE', 'FT BRAGG', 'TROY', 'SHAKER HTS', 'CLEVELAND HTS',
'HAMBURG'], dtype=object)
你也可以试试。如本文档所述，它使用0到n_class-1之间的值对标签进行编码
from sklearn.preprocessing import LabelEncoder
le = LabelEncoder()
df['city_num'] = le.fit_transform(df['city'])
print(df.head())
# city city_num
# 0 LIMA 72
# 1 VACAVILLE 122
# 2 CINCINNATI 21
# 3 GLASGOW 50
# 4 BOWLING GREEN 10
print(len(df.city.unique()))
# 132
print(len(set(df.city_num)))
# 132
from sklearn.preprocessing import OneHotEncoder
ohe = OneHotEncoder()
city_ind = ohe.fit_transform(df.city_num.values.reshape(-1, 1))
print(type(city_ind))
# <class 'scipy.sparse.csr.csr_matrix'>
print(city_ind.shape)
# (132, 132)
print(city_ind[0:2, ].toarray())
# [[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0.]
# [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.
# 0. 0. 0. 0. 0. 0.]]
从sklearn.preprocessing导入OneHotEncoder
ohe=OneHotEncoder（）
city\u ind=ohe.fit\u变换（df.city\u num.values.reformate（-1,1））
印刷品（类型（城市工业））
#
印刷品（城市工业形状）
# (132, 132)
打印（城市索引[0:2，].toarray（））
# [[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0.]
# [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.
# 0. 0. 0. 0. 0. 0.]]
print (df)
city city1
0 a 0
1 b 1
2 c 2
3 d 3
4 d 3
5 f 4
from sklearn.preprocessing import LabelEncoder
le = LabelEncoder()
df['city_num'] = le.fit_transform(df['city'])
print(df.head())
# city city_num
# 0 LIMA 72
# 1 VACAVILLE 122
# 2 CINCINNATI 21
# 3 GLASGOW 50
# 4 BOWLING GREEN 10
print(len(df.city.unique()))
# 132
print(len(set(df.city_num)))
# 132
from sklearn.preprocessing import OneHotEncoder
ohe = OneHotEncoder()
city_ind = ohe.fit_transform(df.city_num.values.reshape(-1, 1))
print(type(city_ind))
# <class 'scipy.sparse.csr.csr_matrix'>
print(city_ind.shape)
# (132, 132)
print(city_ind[0:2, ].toarray())
# [[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0.]
# [ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
# 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.
# 0. 0. 0. 0. 0. 0.]]