HIT-TMG

/

Uni-MoE-2.0-Thinking

like 2

Follow

Lychee Team 90

• "uni_moe.train.grin_moe.data.TrainingArguments",
• "transformers.trainer_utils.SchedulerType",
• "accelerate.state.PartialState",
• "torch.device",
• "torch.bfloat16",
• "transformers.integrations.deepspeed.HfDeepSpeedConfig",
• "transformers.integrations.deepspeed.HfTrainerDeepSpeedConfig",
• "transformers.training_args.OptimizerNames",
• "transformers.trainer_utils.IntervalStrategy",
• "transformers.trainer_pt_utils.AcceleratorConfig",
• "accelerate.utils.dataclasses.DistributedType",
• "transformers.trainer_utils.HubStrategy",
• "accelerate.utils.dataclasses.DeepSpeedPlugin"

How to fix it?