Bad input shape error on SVM training using scikit

Question

I m a little new to scikit and ML. I m trying to train an SVM classifier for one vs all classification. I m using the following code.

g=list()
for i in range(0,120):
    g.append(1)
for i in range(120,240):
    g.append(2)

u=set(g)
numclasses=len(u)

lin_clf = svm.LinearSVC()
lin_clf.fit(features,u)

Features is a 72900*120 array. I m getting features from a different python code and calling that here. It throws the following warning and error.

/usr/lib/python2.7/dist-packages/scipy/misc/pilutil.py:279: 
DeprecationWarning: fromstring() is deprecated. Please call frombytes() instead.
image = Image.fromstring(mode, shape, strdata)

error

ValueError: bad input shape ()

Please comment if you need the code for feature extraction. Thank you in advance.

Solution

Which line of the code is throwing the error? Is it lin_clf.fit(features,u)?

According to the documentation of LinearSVC, the arguments for fit(X,y) are

X : {array-like, sparse matrix}, shape = [n_samples, n_features]

Training vector, where n_samples in the number of samples and n_features is the number of features.

y : array-like, shape = [n_samples]

Target vector relative to X

However, the u in your code is a python set. It should be a numpy array of length 72900.