Get a classification accuracy
def get_classification_accuracy(model, data, true_labels):
# First get the predictions
## YOUR CODE HERE
prds = model.predict(data, output_type='margin')
print prds
# Compute the number of correctly classified examples
## YOUR CODE HERE
correct=0
for i in range(0,len(prds)):
if(prds[i]>=0 and true_labels[i]==1):
correct+=1
elif(prds[i]<0 and true_labels[i]==-1):
correct+=1
# Then compute accuracy by dividing num_correct by total number of examples
## YOUR CODE HERE
accuracy=float(correct)/float(len(true_labels))
return accuracy