Evaluation Loader for DeepSpeed Chat Example step 2 (reward model training)

[harveyp123]

Should the following line be removed? Since there is already eval_sampler initialized in previous few lines:
https://github.com/microsoft/DeepSpeedExamples/blob/dafeb2b3be3a085214faa2f59a8979c051424938/applications/DeepSpeed-Chat/training/step2_reward_model_finetuning/main.py#L236