Closed YananGu closed 3 years ago
May I know more about your environment setup, like torch version and higher version? This could be related to a known issue #2.
The version of Pytorch is 1.4.0 and Higher is 0.2.1. When I set the number of GPU to 1, it works.
Higher does not seem to support multiple GPU training, referring to the issue discussed here.
Got it, thanks for your help!
Hi : when I run "python main.py --cfg ./config/CIFAR10_LT/balms_imba200.yaml", I encountered the following error:
Traceback (most recent call last): File "/data2/gyn/PycharmProjects/BalancedMetaSoftmax-Classification-main-ori/main.py", line 160, in
training_model.train()
File "/data2/gyn/PycharmProjects/BalancedMetaSoftmax-Classification-main-ori/run_networks.py", line 348, in train
self.meta_forward(inputs, labels, verbose=step % self.training_opt['display_step'] == 0)
File "/data2/gyn/PycharmProjects/BalancedMetaSoftmax-Classification-main-ori/run_networks.py", line 272, in meta_forward
trainoutputs, = fmodel(features.detach())
File "/home/gyn/anaconda2/envs/BalanceSoftmax/lib/python3.6/site-packages/torch/nn/modules/module.py", line 532, in call
result = self.forward(*input, kwargs)
File "/home/gyn/anaconda2/envs/BalanceSoftmax/lib/python3.6/site-packages/higher/patch.py", line 460, in _patched_forward
output = self.boxed_forward(*args, *kwargs)
File "/home/gyn/anaconda2/envs/BalanceSoftmax/lib/python3.6/site-packages/higher/patch.py", line 387, in patched_forward
return true_forward(self, args, kwargs)
File "/home/gyn/anaconda2/envs/BalanceSoftmax/lib/python3.6/site-packages/torch/nn/parallel/data_parallel.py", line 151, in forward
replicas = self.replicate(self.module, self.device_ids[:len(inputs)])
File "/home/gyn/anaconda2/envs/BalanceSoftmax/lib/python3.6/site-packages/torch/nn/parallel/data_parallel.py", line 156, in replicate
return replicate(module, device_ids, not torch.is_grad_enabled())
File "/home/gyn/anaconda2/envs/BalanceSoftmax/lib/python3.6/site-packages/torch/nn/parallel/replicate.py", line 146, in replicate
setattr(replica, key, Parameter(param))
File "/home/gyn/anaconda2/envs/BalanceSoftmax/lib/python3.6/site-packages/higher/patch.py", line 279, in setattr
self.update_params(fast_params)
File "/home/gyn/anaconda2/envs/BalanceSoftmax/lib/python3.6/site-packages/torch/nn/modules/module.py", line 576, in getattr
type(self).name, name))
AttributeError: 'InnerFunctionalLinear' object has no attribute 'update_params'.
I have no idea about this error , can you give me some help?