[GitHub] fedorzh opened a new issue #9185: Gluon provided ResNet does not get desirable accuracy on CIFAR10

[GitBox <gi...@apache.org>]

fedorzh opened a new issue #9185: Gluon provided ResNet does not get desirable accuracy on CIFAR10
URL: https://github.com/apache/incubator-mxnet/issues/9185
 
 
   The network provided with gluon by using
   ```
   model_name = 'resnet18_v2'
   
   gpu_count = 1
   _ctx_list = [mx.gpu(i) for i in range(gpu_count)]
   net = gluon.model_zoo.vision.get_model(model_name, pretrained=False, classes=len(np.unique(y)), ctx=_ctx_list)
   ```
   
   does not get desirable accuracy, in comparison with the ResNet defined here:
   https://github.com/tornadomeet/ResNet/blob/master/symbol_resnet.py
   
   with the same optimizer and initialization parameters, even though both claim to reproduce the results.

----------------------------------------------------------------
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