Question

我是数据分析和寻求帮助的新手。我正在使用python从头开始创建Knn算法。我认为我的数据（训练和测试）有问题。我认为我必须转换为浮点数，但我不确定100％。我知道我的函数可以正常工作，因为我尝试了另一个数据集。

from scipy.io import arff
from io import StringIO
import scipy
import pandas as pd
import numpy as np
import math
data_train = scipy.io.arff.loadarff('train.arff')
train = pd.DataFrame(data_train[0])
train.head()
data_test = scipy.io.arff.loadarff('test1.arff') 
print(data_test)
test = pd.DataFrame(data_test[0])
test.head()

from sklearn.cross_validation import train_test_split

X_train, X_test, y_train, y_test = train_test_split(train, test, test_size = 0.1, random_state=42)
print(X_train, X_test, y_train, y_test)

def distance(testpoint, trainpoint):
    # distance between testpoint and trainpoint.
    dist = np.sqrt(np.sum(np.power(testpoint-trainpoint, 2))) 
    return dis

def getNeighbors(X_train, y_train, X_test, k):
        #For each point in X_test, calculate its distance from itself and each point in X_train
        k_neighbors_with_labels = [] # this will be a list (for each test point) of list (contains the tuple (distance,label) of k nearest neighbors). 
        for testpoint in X_test:
            distances_label = [] # this list carries distances between the testpoint and train point
            for (trainpoint,y_train_label) in zip(X_train,y_train):
                # calculate the distance and append it to a distances_label with the associated label.
                distances_label.append((distance(testpoint, trainpoint), y_train_label))
            k_neighbors_with_labels += [sorted(distances_label)[0:k]] # sort the distances and taken the first k neighbors
        return k_neighbors_with_labels
ne = getNeighbors(X_train, y_train, X_test, k = 3)
print(ne)

TypeError                                 Traceback (most recent call last)
<ipython-input-56-3b2868d1fd43> in <module>()
----> 1 ne = getNeighbors(X_train, y_train, X_test, k = 3)
      2 print(ne)

<ipython-input-55-75b4da86d04e> in getNeighbors(X_train, y_train, X_test, k)
      6             for (trainpoint,y_train_label) in zip(X_train,y_train):
      7                 # calculate the distance and append it to a distances_label with the associated label.
----> 8                 distances_label.append((distance(testpoint, trainpoint), y_train_label))
      9             k_neighbors_with_labels += [sorted(distances_label)[0:k]] # sort the distances and taken the first k neighbors
     10         return k_neighbors_with_labels

<ipython-input-42-03d38977fec4> in distance(testpoint, trainpoint)
      1 def distance(testpoint, trainpoint):
      2     # distance between testpoint and trainpoint.
----> 3     dist = np.sqrt(np.sum(np.power(testpoint-trainpoint, 2)))
      4     return distance

TypeError: unsupported operand type(s) for -: 'str' and 'str'

Answer 1

正如评论所提到的-测试点和训练点似乎是字符串。要确认这一点，您可以添加print(type(testpoint))和print(type(trainpoint)) 到您的代码，以找出它们实际上是什么类型。如果它们确实是字符串（错误提示）；假设它们是存储为字符串的数字，那么您可以通过执行以下操作将其转换为int或float：

dist = np.sqrt(np.sum(np.power(float(testpoint)-float(trainpoint), 2)))

根据需要，用float替换int。

有很多方法可以解决此问题，但是根本的问题是，不能像错误所指出的那样对字符串使用-运算符。

-：'str'和'str'

1 个答案: