shikiw

/

LLaVA-v1.5-MoCa-7B-pretrain

like 1

• "llava.train.train.TrainingArguments",
• "transformers.training_args.OptimizerNames",
• "transformers.trainer_utils.SchedulerType",
• "accelerate.state.PartialState",
• "accelerate.utils.dataclasses.DeepSpeedPlugin",
• "transformers.trainer_utils.IntervalStrategy",
• "accelerate.utils.dataclasses.DistributedType",
• "transformers.integrations.deepspeed.HfTrainerDeepSpeedConfig",
• "torch.device",
• "transformers.trainer_utils.HubStrategy",
• "torch.bfloat16"

How to fix it?