data, targets = Data.data() print "training data: ", len(data) test = Data.test() print "test data: ", len(test) data = data + test print "all data: ", len(data) # preprocessing start = time() matrix = BlackboxPreprocess.to_matrix(data) print matrix.shape matrix = BlackboxPreprocess.scale(matrix) #matrix = BlackboxPreprocess.polynomial(matrix, 2) matrix = preprocessing.normalize(matrix, norm='l2') min_max_scaler = preprocessing.MinMaxScaler(feature_range=(-1.,1.)) matrix = min_max_scaler.fit_transform(matrix) #matrix = BlackboxPreprocess.norm(matrix) print matrix.shape data = matrix.tolist() # split training and test data test_data = data[1000:] data, targets = data[:1000], targets[:1000] # testing preds = Classifier.preds(data, targets, test_data, []) print "Duration: ", time() - start submission(preds)
#data = data + psuedo_data #targets = targets + psuedo_targets # preprocessing start = time() print len(data) print len(data[-1]) matrix = BlackboxPreprocess.to_matrix(data) print "(examples, dimensions): ", matrix.shape matrix = BlackboxPreprocess.scale(matrix) matrix = BlackboxPreprocess.polynomial(matrix, 2) #matrix = preprocessing.normalize(matrix, norm='l2') #min_max_scaler = preprocessing.MinMaxScaler(feature_range=(0.,1.)) #matrix = min_max_scaler.fit_transform(matrix) #matrix = BlackboxPreprocess.norm(matrix) print "(examples, dimensions): ", matrix.shape data = matrix.tolist() # split training and CV data cv_data, cv_targets = data[:500], targets[:500] #data, targets = data[500:], targets[500:] # include extra data data, targets = data[500:1000], targets[500:1000] # don't include extra data extra = data # testing preds = Classifier.preds(data, targets, cv_data, cv_targets, extra) s = score(preds, cv_targets, debug=False) print "Score: ", s log(s) print "Duration: ", time() - start