import numpy as np
from sklearn import svm

def get_one_hot(targets, nb_classes):
    res = np.eye(nb_classes)[np.array(targets).reshape(-1)]
    return res.reshape(list(targets.shape)+[nb_classes])



data_slice = 5

def load():
    raw_data = np.loadtxt('wine.data', delimiter=',')
    data_size = len(raw_data)
    np.random.shuffle(raw_data)
    data = np.array_split(raw_data, data_slice)
    ret = [np.split(i, [1,], axis=1) for i in data]
    return ret

data_pieces = load()
val_idx = 0

train_label = np.concatenate(
  [data_pieces[i][0] for i in list(range(data_slice))[:val_idx] + list(range(data_slice))[val_idx+1:]]
)
train_label = train_label - 1
train_label = train_label.flatten().astype(int)

train_data = np.concatenate(
  [data_pieces[i][1] for i in list(range(data_slice))[:val_idx] + list(range(data_slice))[val_idx+1:]]
)

val_label = data_pieces[val_idx][0]
val_data = data_pieces[val_idx][1]
val_label = val_label - 1
val_label = val_label.flatten().astype(int)

clf = svm.SVC(gamma='scale')
clf.fit(train_data, train_label)
print('val_acc', sum(clf.predict(val_data) == val_label) / len(val_label))