brightmart / text_classification

all kinds of text classification models and more with deep learning
MIT License
7.83k stars 2.57k forks source link

seq2seq attention in train with "multi_label_flag=False" #135

Open Tian14267 opened 4 years ago

Tian14267 commented 4 years ago

作者你好,我在运行“a1_seq2seq_attention_train.py”的时候,设置"multi_label_flag=False",却遇到错误提醒:“You must feed a value for placeholder tensor 'decoder_input' with dtype int32 and shape [?,6]” 在你代码 103 、104行说明,使用FLAGS.multi_label_flag=Flase的时候,不需要 decoder_input 才对。 image 请问这是什么情况?