You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2018/07/06 06:13:26 UTC
[GitHub] kwanking opened a new issue #11576: AttributeError: 'ResNetV2'
object has no attribute 'save_parameters'
kwanking opened a new issue #11576: AttributeError: 'ResNetV2' object has no attribute 'save_parameters'
URL: https://github.com/apache/incubator-mxnet/issues/11576
command:
python image_classification.py --dataset mnist --gpus '0' --epochs 2 --mode hybrid --model resnet18_v2 --batch-size 1024 --log-interval 10 --dtype float16
output:
INFO:root:Starting new image-classification task:, Namespace(batch_norm=False, batch_size=1024, builtin_profiler=0, data_dir='', dataset='mnist', dtype='float16', epochs=2, gpus='0', kvstore='device', log_interval=10, lr=0.1, lr_factor=0.1, lr_steps='30,60,90', mode='hybrid', model='resnet18_v2', momentum=0.9, num_workers=4, prefix='', profile=False, resume='', save_frequency=10, seed=123, start_epoch=0, use_pretrained=False, use_thumbnail=False, wd=0.0001)
[23:05:30] src/io/iter_mnist.cc:113: MNISTIter: load 60000 images, shuffle=1, shape=[1024,1,28,28]
[23:05:30] src/io/iter_mnist.cc:113: MNISTIter: load 10000 images, shuffle=1, shape=[1024,1,28,28]
[23:05:31] src/operator/nn/./cudnn/./cudnn_algoreg-inl.h:107: Running performance tests to find the best convolution algorithm, this can take a while... (setting env variable MXNET_CUDNN_AUTOTUNE_DEFAULT to 0 to disable)
INFO:root:Epoch[0] Batch [9] Speed: 7461.372201 samples/sec accuracy=0.803906, top_k_accuracy_5=0.934668
INFO:root:Epoch[0] Batch [19] Speed: 7547.676012 samples/sec accuracy=0.884668, top_k_accuracy_5=0.966650
INFO:root:Epoch[0] Batch [29] Speed: 7268.395011 samples/sec accuracy=0.914486, top_k_accuracy_5=0.977637
INFO:root:Epoch[0] Batch [39] Speed: 7499.912333 samples/sec accuracy=0.930029, top_k_accuracy_5=0.983057
INFO:root:Epoch[0] Batch [49] Speed: 6947.177267 samples/sec accuracy=0.940586, top_k_accuracy_5=0.986445
INFO:root:[Epoch 0] training: accuracy=0.946609, top_k_accuracy_5=0.988298
INFO:root:[Epoch 0] time cost: 11.676347
INFO:root:[Epoch 0] validation: accuracy=0.986220, top_k_accuracy_5=1.000000
Traceback (most recent call last):
File "image_classification.py", line 297, in <module>
main()
File "image_classification.py", line 281, in main
train(opt, context)
File "image_classification.py", line 250, in train
save_checkpoint(epoch, val_acc[0], best_acc)
File "image_classification.py", line 187, in save_checkpoint
net.save_parameters(fname)
AttributeError: 'ResNetV2' object has no attribute 'save_parameters'
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
users@infra.apache.org
With regards,
Apache Git Services