Created by: colin1988

模型配置中output输出多个loss，代码如下：

seq_loss = cross_entropy(input = seq_label, label = target)
crf_l = crf_layer(
       name='crf',
         size=label_dict_len,
         input=feat,
         label=target,
         param_attr=ParameterAttribute(
             name='crfw', learning_rate=mix_hidden_lr))

outputs(seq_loss, crf_l)

训练日志中输出单个loss Batch=3990 samples=598500 AvgCost=1.87198 CurrentCost=1.91031 Eval: sum_evaluator_0=0.0453391 CurrentEval: sum_evaluator_0=0.0440848

请问log输出的单个loss是如何通过outputs中设置的多个loss计算出来的，优化又是如何进行的？