python3でpickleを使ってCIFAR１０を読み込むと「UnicodeDecodeError」...

Question

ChainerでCIFAR-10の分類をしようと思ったのですが、データを読み込むところでエラーが発生して調べても解決策がわからなかったので質問させていただきます。

http://nonbiri-tereka.hatenablog.com/entry/2015/06/21/220506
http://www.cs.toronto.edu/~kriz/cifar.html
上記のサイトを参考に、このようなコードを実行すると、

import pickle
import numpy as np

def unpickle(file):
    fo = open(file, 'rb')
    dict = pickle.load(fo)
    fo.close()
    return dict

X_train = None
y_train = []

for i in range(1,6):
    data_dic = unpickle("cifar-10-batches-py/data_batch_{}".format(i))
    if X_train == None:
        X_train = data_dic['data']
    else:
        X_train = np.vstack((X_train, data_dic['data']))
    y_train += data_dic['labels']

test_data_dic = unpickle("cifar-10-batches-py/test_batch")
x_test = test_data_dic['data']
x_test = x_test.reshape(len(x_test),3,32,32)
y_test = np.array(test_data_dictionary['labels'])
x_train = x_train.reshape((len(x_train),3, 32, 32))
y_train = np.array(y_train)

以下のようなエラーがでます。

In []: %run CIFAR10.py
---------------------------------------------------------------------------
UnicodeDecodeError                        Traceback (most recent call last)
/home/tomoya/CIFAR10.py in <module>()
     14 
     15 for i in range(1,6):
---> 16     data_dic = unpickle("cifar-10-batches-py/data_batch_{}".format(i))
     17     if X_train == None:
     18         X_train = data_dic['data']

/home/tomoya/CIFAR10.py in unpickle(file)
      6 def unpickle(file):
      7     fo = open(file, 'rb')
----> 8     dict = pickle.load(fo)
      9     fo.close()
     10     return dict

UnicodeDecodeError: 'ascii' codec can't decode byte 0x8b in position 6: ordinal not in range(128)

環境は
・ Ubuntu 14.04
・ python 3.5.1 (Anaconda 2.5.0)

解決策もしくは、他の方法でデータを読み込む方法を教えてください！

よろしくお願いします。

Answer 1

import sys
import pickle
import numpy as np

def unpickle(file):
    fp = open(file, 'rb')
    if sys.version_info.major == 2:
        data = pickle.load(fp)
    elif sys.version_info.major == 3:
        data = pickle.load(fp, encoding='latin-1')
    fp.close()

    return data

X_train = None
y_train = []

for i in range(1,6):
    data_dic = unpickle("cifar-10-batches-py/data_batch_{}".format(i))
    if i == 1:
        X_train = data_dic['data']
    else:
        X_train = np.vstack((X_train, data_dic['data']))
    y_train += data_dic['labels']

test_data_dic = unpickle("cifar-10-batches-py/test_batch")
X_test = test_data_dic['data']
X_test = X_test.reshape(len(X_test),3,32,32)
y_test = np.array(test_data_dic['labels'])
X_train = X_train.reshape((len(X_train),3, 32, 32))
y_train = np.array(y_train)

このように変更したら成功しました！