ValueError：预期的 2D 数组，得到 1D 数组：array=[19. 27.896 0. 1. 0. ]。 使用 array.reshape(-1, 1) 重塑数据

Question

I am trying to predict the insurance cost of an individual in the United States using K Nearest Neighbors

I have succesfully trained the data set and it started trowing the error when i tried to predict

ValueError: Expected 2D array, got 1D array instead:
array=[19.    27.896  0.     1.     0.   ].
Reshape your data either using array.reshape(-1, 1) if your data has a single feature or array.reshape(1, -1) if it contains a single sample.

while while executing the following code:

import pandas as pd 
import numpy as np
from sklearn.neighbors import KNeighborsClassifier
knn = KNeighborsClassifier(n_neighbors=1)


insurance = pd.read_csv(r"C:\Users\USER\Desktop\data set\insurance.csv")
insurance2 = pd.read_csv(r"C:\Users\USER\Desktop\data set\insurance.csv")


insurance['charges'] = insurance['charges'].astype('int')

sex_s=[]
for i in insurance['sex']:
    if i == "female":
        sex_s.append(1)
    elif i == "male":
        sex_s.append(0)
    
insurance2['sx']=sex_s

smk=[]
for y in insurance['smoker']:
    if y == "yes":
        smk.append(1)
    elif y == "no":
        smk.append(0)

insurance2['smks']=smk



insurance2 = insurance2.drop(['charges',"region", 'sex','smoker'], axis =1)

insurance2.rename(columns={'smks':'smoker', 'sx':'sex'}, inplace = True)


insurance2['bmi'] = insurance2['bmi'].astype('int')
insurance2['children'] = insurance2['children'].astype('int')
insurance2['sex'] = insurance2['sex'].astype('int')
insurance2['smoker'] = insurance2['smoker'].astype('int')
insurance2['age'] = insurance2['age'].astype('int')
insurance2.dtypes


X = insurance2
y = insurance['charges']


knn.fit(X,y)

knn.predict([19, 27.896, 0, 1, 0])

this is my first time trying to predict

Answer 1

I think you are just missing an extra pair of square brackets on the final line, if you have a look in the docs ( https://scikit-learn.org/stable/modules/generated/sklearn.neighbors.KNeighborsClassifier.html ) you will see they use predict with the following:

neigh.predict([[1.1]])

So just replace

knn.predict([19, 27.896, 0, 1, 0])

with

knn.predict([[19, 27.896, 0, 1, 0]])

For future reference, make sure to look up your question first to check it doesn't exist, as this is a duplicate of Error in Python script "Expected 2D array, got 1D array instead:"? .