无法从检查点恢复模型

作者: seventran创建于 2023年4月21日更新于 2023年5月17日

我无法从检查点恢复训练模型。错误日志如下:/data/zmining/jupyter-notebook/antnh/embeddings/tml/common/checkpointing/snapshot.py,第67行,在restore中:File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torchsnapshot/snapshot.py",第406行,在restore中:self._load_stateful( self.state) #check File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torchsnapshot/snapshot.py",第671行,在_load_stateful中:stateful.load_state_dict(state_dict) File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torch/nn/modules/module.py",第2027行,在load_state_dict中:load(self, state_dict) File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torch/nn/modules/module.py",第2015行,在load中:load(child, child_state_dict, child_prefix) File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torch/nn/modules/module.py",第2015行,在load中:load(child, child_state_dict, child_prefix) File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torch/nn/modules/module.py",第2015行,在load中:load(child, child_state_dict, child_prefix) [上一行重复了2次] File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torch/nn/modules/module.py",第2009行,在load_from_state_dict中:module._load_from_state_dict( File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torch/nn/modules/module.py",第1909行,在_load_from_state_dict中:hook(state_dict, prefix, local_metadata, strict, missing_keys, unexpected_keys, error_msgs) File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torch/nn/modules/module.py",第69行,在__call__中:return self.hook(module, *args, **kwargs) File "/home/zdeploy/anaconda3/envs/quanhm_torchrec/lib/python3.8/site-packages/torchrec/distributed/embeddingbag.py",第439行,在_pre_load_state_dict_hook中:local_shards = state_dict[key].local_shards() KeyError: 'model._dmp_wrapped_module.module.large_embeddings.ebc.embedding_bags.user.weight'

内容来源: twitter/the-algorithm-ml