cornell-zhang / dnn-quant-ocs Goto Github PK

Hi, thank you for sharing the code.
I was wondering whether the quantization method used waas "per-channel" or "per-layer".
I am having trouble looking up the setup in your paper.

Thank you in advance.

Best regards.

Hi authors, I have a problem while inference ResNet50 with the ImageNet test set.
When I inference ResNet50 with the ImageNet test set, I got a strange result that the Top-1 score is 0.102.
However, if I inference ResNet50 with the ImageNet train set, the Top-1 score that I got is about 75.50.
As mentioned above, it's quite strange during inference with the test set.
I've trained it for 50 epochs, but the result told me that the model didn't learn anything from training, it's guessing.
The following picture with blue values is the result I got from the train set, the other is the result I got from the test set.
Hope you can give me some advice to fix this strange situation, thanks.

The result from the train set:

The result from the test set:

Traceback (most recent call last):
File "compress_classifier.py", line 765, in
main()
File "compress_classifier.py", line 298, in main
return evaluate_model(model, criterion, train_loader, test_loader, pylogger, args)
File "compress_classifier.py", line 701, in evaluate_model
top1, _, _ = test(test_loader, model, criterion, loggers, args=args)
File "compress_classifier.py", line 483, in test
return _validate(test_loader, model, criterion, loggers, args)
File "compress_classifier.py", line 515, in _validate
output = model(inputs)
File "/home/songyan3/anaconda3/envs/python3.6/lib/python3.6/site-packages/torch/nn/modules/module.py", line 491, in call
result = self.forward(*input, **kwargs)
File "/home/songyan3/anaconda3/envs/python3.6/lib/python3.6/site-packages/torch/nn/parallel/data_parallel.py", line 114, in forward
outputs = self.parallel_apply(replicas, inputs, kwargs)
File "/home/songyan3/anaconda3/envs/python3.6/lib/python3.6/site-packages/torch/nn/parallel/data_parallel.py", line 124, in parallel_apply
return parallel_apply(replicas, inputs, kwargs, self.device_ids[:len(replicas)])
File "/home/songyan3/anaconda3/envs/python3.6/lib/python3.6/site-packages/torch/nn/parallel/parallel_apply.py", line 65, in parallel_apply
raise output
File "/home/songyan3/anaconda3/envs/python3.6/lib/python3.6/site-packages/torch/nn/parallel/parallel_apply.py", line 41, in _worker
output = module(*input, **kwargs)
File "/home/songyan3/anaconda3/envs/python3.6/lib/python3.6/site-packages/torch/nn/modules/module.py", line 491, in call
result = self.forward(*input, **kwargs)
File "/home/songyan3/anaconda3/envs/python3.6/lib/python3.6/site-packages/torchvision/models/resnet.py", line 139, in forward
x = self.conv1(x)
File "/home/songyan3/anaconda3/envs/python3.6/lib/python3.6/site-packages/torch/nn/modules/module.py", line 491, in call
result = self.forward(*input, **kwargs)
File "/home/songyan3/experiments/dnn-quant-ocs/distiller/quantization/ocs.py", line 169, in forward
assert(self.profile_info)

Excuse me, have you met such question in running example.sh? it seems that self.profile_info has not been assigned =A=