NorthFun

raw：”找不到背景和滤镜“ DDParser: ``` text:找不到, pos:v, dep:HED, rel:0 text:背景, pos:n, dep:ATT, rel:4 text:和, pos:c, dep:MT, rel:4 text:滤镜, pos:n, dep:VOB, rel:1 ``` "背景"和"滤镜"应该是并列关系CC。 spacy可以正确解析： ``` text:找, pos:VERB, dep:ROOT, head:找 text:不, pos:ADV, dep:dep,...

fix typo

By the way, what are formulas about `point-wise mutual information, point-wise KL divergence` you used to calculate n-gram's(n>2) phrase quality? Thx~

epoch: 10 train_batch_size: 8 evaluation_batch_size: 16 learning_rate: 3e-5 预训练模型：bert_base_chinese 自己的中文数据集 bert+softmax: acc: 0.0729 - recall: 0.4249 - f1: 0.1245 - loss: 1.0244 bert+crf: acc: 0.4268 - recall: 0.5729 - f1:...