CODEWITHSUNDEEP
pythonnumpyscikit-learn
Roman
Roman

Reputation: 3241

Value Error while running SVM in Sklearn

I have the following problem of doing support vector machine with numpy arrays.

import numpy as np
from sklearn import svm

I have 3 classes/labels (male, female, na), denoted as follows:

labels = [0,1,2]

Each class was defined by 3 variables (height, weight, age) as the training data:

male_height = np.array([111,121,137,143,157])
male_weight = np.array([60,70,88,99,75])
male_age = np.array([41,32,73,54,35])

males = np.hstack([male_height,male_weight,male_age])

female_height = np.array([91,121,135,98,90])
female_weight = np.array([32,67,98,86,56])
female_age = np.array([51,35,33,67,61])

females = np.hstack([female_height,female_weight,female_age])

na_height = np.array([96,127,145,99,91])
na_weight = np.array([42,97,78,76,86])
na_age = np.array([56,35,49,64,66])

nas = np.hstack([na_height,na_weight,na_age])

Now I have to fit the support vector machine method for the training data to predict the class given those three variables:

height_weight_age = [100,100,100]

clf = svm.SVC()
trainingData = np.vstack([males,females,nas])

clf.fit(trainingData, labels)

result = clf.predict(height_weight_age)

print result

Unfortunately, the following error occurs:

  ValueError: X.shape[1] = 3 should be equal to 15, the number of features at training time

How should I modify the trainingData and labels to get the correct answer?

Upvotes: 0

Views: 2120

Answers (1)

Fred Foo
Fred Foo

Reputation: 363817

hstack gives 1-d arrays. You need 2-d arrays of shape (n_samples, n_features), which you can get from vstack.

In [7]: males = np.hstack([male_height,male_weight,male_age])

In [8]: males
Out[8]: 
array([111, 121, 137, 143, 157,  60,  70,  88,  99,  75,  41,  32,  73,
        54,  35])

In [9]: np.vstack([male_height,male_weight,male_age])
Out[9]: 
array([[111, 121, 137, 143, 157],
       [ 60,  70,  88,  99,  75],
       [ 41,  32,  73,  54,  35]])

In [10]: np.vstack([male_height,male_weight,male_age]).T
Out[10]: 
array([[111,  60,  41],
       [121,  70,  32],
       [137,  88,  73],
       [143,  99,  54],
       [157,  75,  35]])

You also need to pass a list/array of labels that reflects the label of each sample, rather than just enumerating the labels that exist. After fixing all of your variables, I can train an SVM and apply it as follows:

In [19]: clf = svm.SVC()

In [20]: y = ["male"] * 5 + ["female"] * 5 + ["na"] * 5

In [21]: X = np.vstack([males, females, nas])

In [22]: clf.fit(X, y)
Out[22]: 
SVC(C=1.0, cache_size=200, class_weight=None, coef0=0.0, degree=3, gamma=0.0,
  kernel='rbf', max_iter=-1, probability=False, random_state=None,
  shrinking=True, tol=0.001, verbose=False)

In [23]: height_weight_age = [100,100,100]

In [24]: clf.predict(height_weight_age)
Out[24]: 
array(['female'], 
      dtype='|S6')

(Note that I'm using string labels instead of numeric ones. I'd also advise you standardize the feature values, since they have rather different ranges.)

Upvotes: 2

Related Questions