The bretschneideri from notcome

bretschneideri's People

Contributors

Watchers

bretschneideri's Issues

TypeError: type object argument after ** must be a mapping, not Namespace

Some tensors are not on GPU.

I should have tested on a GPU.

"load" requires "--resuming"

Memory leak occurs when summarizing with differentiable tensors.

We should create gradient-free copies of summarized results.

Overwriting would not work.

Traceback (most recent call last):
  File "bret.py", line 217, in <module>
    'overwriting': True
  File "/home/selveskii/.conda/envs/lign199/lib/python3.7/site-packages/bretschneideri/__init__.py", line 13, in launch
    agent.train()
  File "/home/selveskii/.conda/envs/lign199/lib/python3.7/site-packages/bretschneideri/agent.py", line 123, in train
    self.io.save_torch_states(self.index.total_epoch, self.states_dict)
  File "/home/selveskii/.conda/envs/lign199/lib/python3.7/site-packages/bretschneideri/directory_manager.py", line 53, in save_torch_states
    torch.save(dict[key].state_dict(), path)
  File "/home/selveskii/.conda/envs/lign199/lib/python3.7/site-packages/torch/serialization.py", line 219, in save
    return _with_file_like(f, "wb", lambda f: _save(obj, f, pickle_module, pickle_protocol))
  File "/home/selveskii/.conda/envs/lign199/lib/python3.7/site-packages/torch/serialization.py", line 142, in _with_file_like
    f = open(f, mode)
FileNotFoundError: [Errno 2] No such file or directory: '/home/selveskii/sessions/bret/states/0000.model'

Recommend Projects