xiaoya6666
xiaoya6666
> Hi @xiaoya6666 > > I couldn't achieve the performance of the JMEE paper too. > The following results were printed on my console: > > ``` > python -m...
> @xiaoya6666 I also try to reproduce JMEE,but can't achieve the result of paper. I try to use Bert to replace the word embedding of JMEE ,but ed F1 just...
我没有注意到这个问题啊，我触发词预测能到70%，但是都预测为O预测效果不是会降低么？但我触发词预测的效果还算比较稳定，主要是论元提不上去 2019-12-31 lxyyy123 发件人：Hanlard 发送时间：2019-12-31 11:20 主题：Re: [nlpcl-lab/bert-event-extraction] have you tried to use bert to improve the performance of JMEE? (#4) 收件人："nlpcl-lab/bert-event-extraction" 抄送："xiaoya6666","Mention" 我观察了一下, 好像是因为, step(batch_size=8)大于20之后, 触发词都预测为"O"了, 然后loss一直等于trigger_loss, argument_loss就不再下降了, 角色识别效果就提不上去了,...
现在bert这个代码就是参考JMEE的论元抽取方法的吧。emmm 2020-01-02 lxyyy123 发件人：Hanlard 发送时间：2020-01-02 11:16 主题：Re: [nlpcl-lab/bert-event-extraction] have you tried to use bert to improve the performance of JMEE? (#4) 收件人："nlpcl-lab/bert-event-extraction" 抄送："xiaoya6666","Mention" 是我搞错了，抱歉。论元我觉得可以参考一下jmee的模型发自我的华为手机-------- 原始邮件 --------发件人： xiaoya6666 日期： 2019年12月31日周二 下午4:38收件人：...
bert的，我用的是那个bert-base-uncased模型微调的 2020-01-03 lxyyy123 发件人：ScuLilei2014 发送时间：2020-01-02 13:16 主题：Re: [nlpcl-lab/bert-event-extraction] have you tried to use bert to improve the performance of JMEE? (#4) 收件人："nlpcl-lab/bert-event-extraction" 抄送："xiaoya6666","Mention" @xiaoya6666 您好，你的触发词70%是bert的代码还是jmee的，能加下QQ详聊吗(2512156864) — You are receiving this...
> @zijunsun how do you get the result, I just got ed_f1:0.54, ae_f1:0.29 can you improve the code or can you share me a Hyperparameter? hi, I got the ed_f1:55%,...
我发现似乎是他读数据的方式有点问题
同学，你有没有遇到用BucketIterator生成迭代器会生成一个无限循环的迭代器的情况？这个迭代器有问题么？ 2019-12-07 lxyyy123 发件人：lhjner 发送时间：2019-12-04 20:45 主题：Re: [lx865712528/EMNLP2018-JMEE] json.decoder.JSONDecodeError: Expecting value: line 1 column 2 (char 1) (#20) 收件人："lx865712528/EMNLP2018-JMEE" 抄送："xiaoya6666","Comment" 我发现似乎是他读数据的方式有点问题 对，没错，是读取.json文件时的错误，跟.json文件本身没有关系 — You are receiving this because you commented....
> Hello, > > I am trying to reproduce the same results using the same parameters, but when I run your code for some time, the loss keeps going down...
请问您得到的最好的结果是多少啊 @Hanlard