Thank you for publishing this nice work. It reports an error when I run the code at the function 'train()' in training.py, line 663. The following error displayed:
Traceback (most recent call last):
File "D:/cmWorks/backdoor_federated_learning-master/training.py", line 667, in
is_poison=helper.params['is_poison'], last_weight_accumulator=weight_accumulator)
File "D:/cmWorks/backdoor_federated_learning-master/training.py", line 183, in train
name='Classification Loss', win='poison')
File "D:\cmWorks\backdoor_federated_learning-master\models\simple.py", line 43, in trainvis
opts=dict(showlegend=True, width=700, height=400, title='Train loss{0}'.format(self.created_time)))
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom__init.py", line 389, in wrapped_f
return f(*args, *kwargs)
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom__init.py", line 1715, in line
update=update, name=name)
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom\init__.py", line 389, in wrapped_f
return f(args, **kwargs)
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom__init.py", line 1640, in scatter
return self._send(data_to_send, endpoint=endpoint)
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom__init__.py", line 711, in _send
data=json.dumps(msg),
File "E:\Anaconda3\envs\pyTorch\lib\json\init__.py", line 231, in dumps
return _default_encoder.encode(obj)
File "E:\Anaconda3\envs\pyTorch\lib\json\encoder.py", line 199, in encode
chunks = self.iterencode(o, _one_shot=True)
File "E:\Anaconda3\envs\pyTorch\lib\json\encoder.py", line 257, in iterencode
return _iterencode(o, 0)
File "E:\Anaconda3\envs\pyTorch\lib\json\encoder.py", line 180, in default
o.class.name__)
TypeError: Object of type 'Tensor' is not JSON serializable
It seems like the function "json.dumps()" can not fit the type of some "kwargs". Is it caused by the version's problems or something else?
Thank you for publishing this nice work. It reports an error when I run the code at the function 'train()' in training.py, line 663. The following error displayed:
Traceback (most recent call last): File "D:/cmWorks/backdoor_federated_learning-master/training.py", line 667, in
is_poison=helper.params['is_poison'], last_weight_accumulator=weight_accumulator)
File "D:/cmWorks/backdoor_federated_learning-master/training.py", line 183, in train
name='Classification Loss', win='poison')
File "D:\cmWorks\backdoor_federated_learning-master\models\simple.py", line 43, in trainvis
opts=dict(showlegend=True, width=700, height=400, title='Train loss{0}'.format(self.created_time)))
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom__init.py", line 389, in wrapped_f
return f(*args, *kwargs)
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom__init.py", line 1715, in line
update=update, name=name)
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom\init__.py", line 389, in wrapped_f
return f(args, **kwargs)
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom__init.py", line 1640, in scatter
return self._send(data_to_send, endpoint=endpoint)
File "E:\Anaconda3\envs\pyTorch\lib\site-packages\visdom__init__.py", line 711, in _send
data=json.dumps(msg),
File "E:\Anaconda3\envs\pyTorch\lib\json\init__.py", line 231, in dumps
return _default_encoder.encode(obj)
File "E:\Anaconda3\envs\pyTorch\lib\json\encoder.py", line 199, in encode
chunks = self.iterencode(o, _one_shot=True)
File "E:\Anaconda3\envs\pyTorch\lib\json\encoder.py", line 257, in iterencode
return _iterencode(o, 0)
File "E:\Anaconda3\envs\pyTorch\lib\json\encoder.py", line 180, in default
o.class.name__)
TypeError: Object of type 'Tensor' is not JSON serializable
It seems like the function "json.dumps()" can not fit the type of some "kwargs". Is it caused by the version's problems or something else?
Thanks a lot!