CODEWITHSUNDEEP
pythonmachine-learningscikit-learnsvmmultilabel-classification
michelle.70
michelle.70

Reputation: 183

Scikit-learn multiple targets

I left this example to create a classifier images with scikit-learn.

Though each image belongs to a single category everything works, but each image may belong to several categories, such as: photos with daytime dogs, pictures of cats at night, pictures of cats and dogs at night etc ... I wrote:

target=[[0,1],[0,2],[1,2],[0,2,3]]
target = MultiLabelBinarizer().fit_transform(target)

classifier = svm.SVC(gamma=0.001)
classifier.fit(data, target)

but I get this error:

Traceback (most recent call last):
  File "test.py", line 49, in <module>
    classifier.fit(data, target)
  File "/home/mezzo/.local/lib/python2.7/site-packages/sklearn/svm/base.py", line 151, in fit
    y = self._validate_targets(y)
  File "/home/mezzo/.local/lib/python2.7/site-packages/sklearn/svm/base.py", line 514, in _validate_targets
    y_ = column_or_1d(y, warn=True)
  File "/home/mezzo/.local/lib/python2.7/site-packages/sklearn/utils/validation.py", line 551, in column_or_1d
    raise ValueError("bad input shape {0}".format(shape))
ValueError: bad input shape (4, 4)

Complete code

import numpy as np
import PIL
from PIL import Image
import matplotlib.image as mpimg

# The digits dataset
digits = datasets.load_digits()

def normalize(old_im):
    base = 400

    if (old_im.size[0] > old_im.size[1]):
        wpercent = (base/float(old_im.size[0]))
        hsize = int((float(old_im.size[1])*float(wpercent)))
        old_im = old_im.resize((base,hsize), PIL.Image.ANTIALIAS)
    else:
        wpercent = (base/float(old_im.size[1]))
        wsize = int((float(old_im.size[0])*float(wpercent)))
        old_im = old_im.resize((wsize, base), PIL.Image.ANTIALIAS)

    old_size = old_im.size

    new_size = (base, base)
    new_im = Image.new("RGB", new_size)
    new_im.paste(old_im, ((new_size[0]-old_size[0])/2,
                          (new_size[1]-old_size[1])/2))

    #new_im.show()
    new_im.save('prov.jpg')
    return mpimg.imread('prov.jpg')

# To apply a classifier on this data, we need to flatten the image, to
# turn the data in a (samples, feature) matrix:
imgs = np.array([normalize(Image.open('/home/mezzo/Immagini/1.jpg')),normalize(Image.open('/home/mezzo/Immagini/2.jpg')),normalize(Image.open('/home/mezzo/Immagini/3.jpg')),normalize(Image.open('/home/mezzo/Immagini/4.jpg'))])
n_samples = len(imgs)
data = imgs.reshape((n_samples, -1))

target=[[0,1],[0,2],[1,2],[0,2,3]]
target = MultiLabelBinarizer().fit_transform(target)

# Create a classifier: a support vector classifier
classifier = svm.SVC(gamma=0.001)

# We learn the digits on the first half of the digits
classifier.fit(data, target)

# Now predict the value of the digit on the second half:
predicted = classifier.predict(data)

print("Classification report for classifier %s:\n%s\n"
      % (classifier, metrics.classification_report(target, predicted)))
print("Confusion matrix:\n%s" % metrics.confusion_matrix(target, predicted))

Upvotes: 3

Views: 2648

Answers (2)

ali_m
ali_m

Reputation: 74154

Scikit-learn's SVM implementation doesn't natively support multilabel classification, although it has various other classifiers that do:

It's also possible to do multilabel classification with SVM by treating each unique combination of labels as a separate class. You could simply replace each unique row in your target matrix with a single integer label, which can be done efficiently using np.unique:

d = np.dtype((np.void, target.dtype.itemsize * target.shape[1]))
_, ulabels = np.unique(np.ascontiguousarray(target).view(d), return_inverse=True)

Then you could train SVM as you would for a single-label classification problem:

clf = svm.SVC()
clf.fit(data, ulabels)

One potential caveat is that your classifier's performance may be poor for rare combinations of labels if you don't have a large number of training examples.

Upvotes: 1

Farseer
Farseer

Reputation: 4172

This happens because your target is :

array([[1, 1, 0, 0],
       [1, 0, 1, 0],
       [0, 1, 1, 0],
       [1, 0, 1, 1]])

Your target must be of shape (m,), where m is number of examples. One way to deal with that is to convert your binary arrays to labels, like that:

for item in target:
    print(sum(1<<i for i, b in enumerate(item) if b))

The output of this will be :

3
5
6
13

Now you can use [3,5,6,13] as your targets.

Upvotes: 0

Related Questions