MCPcopy Create free account
hub / github.com/HA-Transformer/MAT / load_state_dict

Method load_state_dict

fairseq/optim/fp16_optimizer.py:125–135  ·  view source on GitHub ↗

Load an optimizer state dict. In general we should prefer the configuration of the existing optimizer instance (e.g., learning rate) over that found in the state_dict. This allows us to resume training from a checkpoint using a new set of optimizer args.

(self, state_dict, optimizer_overrides=None)

Source from the content-addressed store, hash-verified

source not stored for this graph (policy: none)

Callers 1

load_state_dictMethod · 0.45

Calls

no outgoing calls

Tested by

no test coverage detected