alperenenes

/

vlmr1_grpo_less_rewards

like 0

• "_codecs.encode",
• "collections.OrderedDict",
• "numpy._core.multiarray._reconstruct",
• "torch._utils._rebuild_tensor_v2",
• "torch.ByteStorage",
• "numpy.dtype",
• "numpy.ndarray"

How to fix it?

• "_codecs.encode",
• "collections.OrderedDict",
• "numpy._core.multiarray._reconstruct",
• "torch._utils._rebuild_tensor_v2",
• "torch.ByteStorage",
• "numpy.dtype",
• "numpy.ndarray"

How to fix it?

• "open_r1.trainer.grpo_config.GRPOConfig",
• "transformers.trainer_utils.IntervalStrategy",
• "transformers.training_args.OptimizerNames",
• "accelerate.utils.dataclasses.DeepSpeedPlugin",
• "torch.bfloat16",
• "transformers.trainer_utils.SchedulerType",
• "torch.device",
• "transformers.trainer_utils.HubStrategy",
• "accelerate.utils.dataclasses.DistributedType",
• "accelerate.state.PartialState",
• "transformers.trainer_utils.SaveStrategy",
• "transformers.integrations.deepspeed.HfDeepSpeedConfig",
• "transformers.integrations.deepspeed.HfTrainerDeepSpeedConfig",
• "transformers.trainer_pt_utils.AcceleratorConfig"

How to fix it?