Question

我对python / numpy很新，并不完全清楚它。

我一直在尝试实现一种算法，并且在尝试使用其转置获取数组的点积时陷入某一点。我得到的错误是这个

TypeError：只能将整数标量数组转换为标量索引。

以下是我的参考代码。

import pandas as pd
import numpy as np

dataset=pd.read_csv('SCLC_study_output_filtered_2.csv',header=0,delimiter=",")

#forming the first class
class_1 = dataset.iloc[0:20,1:20].values

#forming the second class
class_2 = dataset.iloc[20:41,1:20].values

mean_c1 = np.mean(class_1, axis=0)

#Taking mean of class 2 
mean_c2 = np.mean(class_2, axis=0)

mean_classes =[mean_c1,mean_c2]

#Calculating S-within for class-1
scatter_within_c1 = np.zeros((19,19))
for i in range(0,20):
    for col in class_1:
        col, m = col.reshape(19,1), mean_c1.reshape(19,1)
        sub = np.subtract(col,m)
        scatter_within_c1 += np.prod(sub,np.transpose(sub))

Answer 1

查看docs for np.prod()：

override init () {
    super.init();    // necessary otherwise next line is an error
    pluginInitialize(); 
}
override func pluginInitialize() {
    players = [:];
}

返回给定轴上的数组元素的乘积。

numpy.prod(a, axis=None, dtype=None, out=None, keepdims=<class numpy._globals._NoValue>)函数不适用于两个不同的数组。对于点积，您可以使用np.dot() function或等效地， ndarray.dot() method：

np.prod()

错误：TypeError：只能将整数标量数组转换为标量索引

1 个答案: