RL training error

[HN246]

当我尝试CIDEr优化训练时，遇到报错：“ gen_result, sample_logprobs, core_args = decoder(vrg_data, fc_feats, att_feats, weak_relas, att_masks, opt={'sample_max': 0, 'return_core_args': True}, mode='sample') ValueError: need more than 2 values to unpack” decoder()只能返回gen_result和sample_logprobs，不能返回core_args.

我想问一下，decoder（）的返回值具体由那一步决定？？？