Message382294
torch.nn.functional.nll_loss returns Tensor objects which contain a loss value as well as a grad_fn object. Pickle does not throw an exception when serializing (dumps) the Tensor object. When loading (loads) the serialized data, the grad_fn object is lost and it becomes a requires_grad object.
However, if one attempts to serialize the grad_fn object encapsulated in the Tensor object, Pickle then throws a TypeError (TypeError: cannot pickle 'NllLossBackward' object). This behavior seems inconsistent. Perhaps serialization of NllLossBackward objects should be supported or their encapsulating Tensors should also throw a TypeError.
This behavior should be easily reproducible. |
|
Date |
User |
Action |
Args |
2020-12-02 06:23:16 | dkostecki | set | recipients:
+ dkostecki |
2020-12-02 06:23:16 | dkostecki | set | messageid: <1606890196.44.0.812849435025.issue42530@roundup.psfhosted.org> |
2020-12-02 06:23:16 | dkostecki | link | issue42530 messages |
2020-12-02 06:23:16 | dkostecki | create | |
|