python3.x - python3.5的numpy设置array为float64，报错"name float64 is not defined"

Question

trainWeights = gradDescent(np.array(trainingSet,dtype=float64), trainingLabels)

把dtype设置为float64会报错 NameError: name 'float64' is not defined

但是如果去掉"dtype=float64"会出现warning：DataConversionWarning: Data with input dtype <U3 was converted to float64 by MinMaxScaler.
warnings.warn(msg, DataConversionWarning) 提醒被转换成了float64

请问应该如何显式转换呢？

Answer 1

Is it possible that you didn’t import float64
Of course it’s not right for you to import like this literally. .
Like this. .

import numpy as np
dtype=np.float64

Answer 2

float64 is a class under numpy. If you want to use float64 without modification, you need to import it like this

import numpy as np
from numpy import float64


c = np.array([], dtype=float64)

If there is no from numpy import float64, use it like this

c = np.array([], dtype=np.float64)