ChChwang comments

Results 33 comments of


                                            ChChwang

> 请问这个问题要怎么解决呢？怎么就关闭了啊 batch_size 在py文件里改不行，bert-base-ner-train 执行的时候，-batch_size 16 就可以了

> 请问英文读取数据后 token里面有#是怎么回事例如 cl ##air ##mon ##t 谢谢 ##表示这个词是接着前面的，这样使得WordPiece切分是可逆的——我们可以恢复出“真正”的词

我用的python2，restore时间会长点，但是不会卡主。

具体时间没有计算，可以看看计算资源使用率。

> ASTER is the best algorithm in Chinese Scene Text Recognition. have you try distributed training ? how big of your training datasets?

git clone again and it worked

does the network of left-to-right decoder is same to the right-to-left's ? doing right-to-left decoder by provide_groundtruth reverse?@bgshih

--exp_dir ....

> It failed on Tensorflow 1.13 (works in 1.4). haven't tried any other versions can you run c_ops/build.sh ? tf

yes, the code repeats