Jingjing Gong

主目录下的 nlpcc_data.tar.gz

filter个数设多一点

因为代码重用了vocab类，你不想看到unk可以载入tag词典的时候pop掉unk

你换数据集需要重新做词表以及类别词表也就是id2tag.txt文件

数据集换了id2tag.txt 以及vocab.txt都需要重新根据自己的数据集生成。没有negative samples loss function你可以看代码自己加上就好了。

`pre_trained = True` 才行，不跟新参数请改下面代码： https://github.com/FudanNLP/nlpcc2017_news_headline_categorization/blob/067b54681f218e974c739d5a4f5ed7da26aed277/model.py#L133 这行设置一个trainable=False

I use AdamOptimizer, my first time to have observed a going up training loss, like from 1.2-> 0.4->1.0. And I have no idea why. It seems getting better when I...

batch size set to 32, lr set to 0.0001. Trained like 10 epochs, but the update number is huge since the data is abundant. so according to your plot it's...