Something wrong when I ran the example of image-classification

[Qingmu]

I just installed mxnet in my Banana Pi, and wanted to test it. While I ran the example in image-classification, it didn't work. I cannot find the reason for my first try with mxnet. Does someone know what happens?

When I ran the code (without gup):

python train_cifar10.py --network resnet --num-layers 110 --batch-size 128 --gpus ‘’

I got:

ValueError: invalid literal for int() with base 10: '\xe2\x80\x98\xe2\x80\x99'

While I ran the example:

python train_mnist.py --network mlp

I got:

IOError: CRC check failed 0x6d9b6e54 != 0x85c4e58dL

Environment info

Operating System:
debian 8 in Banana Pi

Package used (Python/R/Scala/Julia):
Python

MXNet commit hash (git rev-parse HEAD):
7907e45

Python version and distribution:
2.7.9

Error Message:

root@armbian:~/mxnet/example/image-classification# python train_cifar10.py --network resnet --num-layers 110 --batch-size 128 --gpus ‘’
INFO:root:start with arguments Namespace(batch_size=128, benchmark=0, data_nthreads=4, data_train='data/cifar10_train.rec', data_val='data/cifar10_val.rec', disp_batches=20, gpus='\xe2\x80\x98\xe2\x80\x99', image_shape='3,28,28', kv_store='device', load_epoch=None, lr=0.1, lr_factor=0.1, lr_step_epochs='200,250', max_random_aspect_ratio=0, max_random_h=36, max_random_l=50, max_random_rotate_angle=0, max_random_s=50, max_random_scale=1, max_random_shear_ratio=0, min_random_scale=1, model_prefix=None, mom=0.9, network='resnet', num_classes=10, num_epochs=300, num_examples=50000, num_layers=110, optimizer='sgd', pad_size=4, random_crop=1, random_mirror=1, rgb_mean='123.68,116.779,103.939', test_io=0, top_k=0, wd=0.0001)
[22:44:45] src/io/iter_image_recordio.cc:221: ImageRecordIOParser: data/cifar10_train.rec, use 1 threads for decoding..
[22:44:47] src/io/iter_image_recordio.cc:221: ImageRecordIOParser: data/cifar10_val.rec, use 1 threads for decoding..
Traceback (most recent call last):
File "train_cifar10.py", line 53, in
fit.fit(args, sym, data.get_rec_iter)
File "/root/mxnet/example/image-classification/common/fit.py", line 131, in fit
mx.gpu(int(i)) for i in args.gpus.split(',')]
ValueError: invalid literal for int() with base 10: '\xe2\x80\x98\xe2\x80\x99'

root@armbian:~/mxnet/example/image-classification# python train_mnist.py --network mlp
INFO:root:start with arguments Namespace(batch_size=64, disp_batches=100, gpus=None, kv_store='device', load_epoch=None, lr=0.1, lr_factor=0.1, lr_step_epochs='10', model_prefix=None, mom=0.9, network='mlp', num_classes=10, num_epochs=20, num_examples=60000, num_layers=None, optimizer='sgd', test_io=0, top_k=0, wd=0.0001)
Traceback (most recent call last):
File "train_mnist.py", line 76, in
fit.fit(args, sym, get_mnist_iter)
File "/root/mxnet/example/image-classification/common/fit.py", line 103, in fit
(train, val) = data_loader(args, kv)
File "train_mnist.py", line 41, in get_mnist_iter
't10k-labels-idx1-ubyte.gz', 't10k-images-idx3-ubyte.gz')
File "train_mnist.py", line 24, in read_data
image = np.fromstring(fimg.read(), dtype=np.uint8).reshape(len(label), rows, cols)
File "/usr/lib/python2.7/gzip.py", line 261, in read
self._read(readsize)
File "/usr/lib/python2.7/gzip.py", line 315, in _read
self._read_eof()
File "/usr/lib/python2.7/gzip.py", line 354, in _read_eof
hex(self.crc)))
IOError: CRC check failed 0x6d9b6e54 != 0x85c4e58dL

[piiswrong]

@mli

[mli]

the first error seems to be --gpus is not given ascii input
the second probably due to the download of mnist is failed, please remove the the mnist data in ./data and try it again

[ysh329]

You can try solution below:

IOError of train_mnist.py :( · Issue #4285 · dmlc/mxnet
#4285

[yajiedesign]

This issue is closed due to lack of activity in the last 90 days. Feel free to reopen if this is still an active issue. Thanks!