Sequential/Partial unpickling and loading of models #97196
Labels
module: pickle
Problems related to pickling of PyTorch objects
triaged
This issue has been looked at a team member, and triaged and prioritized into an appropriate module
馃殌 The feature, motivation and pitch
Given a
90GB
model file that should not be run, but rather just inspected, you might want to load it partially.Example usage:
This is not a suggestion regarding syntax. At the moment, this is hardly possible:
torch.load
will use about90GB
to load the model.pickle.Unpickler
still wants to construct thetorch.storage
's for all tensors, which conceptually should not be neccessary (also, doesn't work as some tensors may share storage)I'm not sure this is an easy patch, or if it even is possible. Please leave any suggestions in the comments.
Alternatives
No response
Additional context
No response
The text was updated successfully, but these errors were encountered: