_pickle.UnpicklingError: pickle data was truncated for loading CIFAR100

Hello Guys,

I am trying to load CIFAR100 using cifar100.load_data() using keras. But I am encountering an error which says:

Using TensorFlow backend.
Traceback (most recent call last):
  File "resnet_cifar_new100.py", line 52, in <module>
    (x_train, y_train), (x_test, y_test) = cifar100.load_data()
  File "/home/bansa01/anaconda3/envs/keras/lib/python3.6/site-packages/keras/datasets/cifar100.py", line 34, in load_data
    x_train, y_train = load_batch(fpath, label_key=label_mode + '_labels')
  File "/home/bansa01/anaconda3/envs/keras/lib/python3.6/site-packages/keras/datasets/cifar.py", line 27, in load_batch
    d = cPickle.load(f, encoding='bytes')
_pickle.UnpicklingError: pickle data was truncated

I have made no changes whatsoever in the respective file. And the version of Keras and Tensorflow I am using are:

(keras) bansa01@vita2:/ssd2/home/bansa01/resnet_final/tmp_spectral_new$ python -c 'import keras; print(keras.__version__)'
Using TensorFlow backend.

2.2.4

(keras) bansa01@vita2:/ssd2/home/bansa01/resnet_final/tmp_spectral_new$ python -c 'import tensorflow; print(tensorflow.__version__)'

1.11.0

Please suggest what should be done.

Regards Nitin

[carloshdez90]

I had the same error, try with this https://stackoverflow.com/a/52496786 it worked for me.

[SGQ-py]

Thank you for your reply, it helped me.

| | wxxd | | [e-mail] | 签名由网易邮箱大师定制 On 10/8/2020 21:55，[e-mail] wrote：

For me the @xscharlie did work. The issue was that the downloaded files were broken, so it couldn’t load it. Erased them and downloaded them again and it worked. But I guess since it didn’t work for @AayushTyagi1 , the issue still remains open.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub, or unsubscribe.