zyh3826
zyh3826
change data to: ``` d = { 1999: [[19990012, '动画片']], 1114: [[11140004, '动画片'], [11140011, '冒险']], 1101: [[11010020, '冒险']] } also, get the wrong output 1999->冒险->11010020 1999->动画片->11140004 1114->冒险->11010020 1114->动画片->11140004 1101->冒险->11010020 1101->动画片->11140004...
> Does this model apply to Chinese data？ hi, I'm applying this model(fine tune bert base) to Chinese OneNote data, using a pytorch version, but I got a very low...
您好，我的roformer版本是0.3.0，transformers是4.12.5
> 我的正常，你可以重新装一下这两个库。  重装了，还是不行，我在docker里也搞了，一样出错
插楼提问，博主您好，请问ontonotes-release-5.0和conll-2012数据集有啥区别吗
did you solve this problem?
+1 to this issue, does this problem fix? cuml version 0.19 in docker, cuda10.2, my dataset: https://drive.google.com/file/d/1P3UcuCrO6FU_NKv4E59goDzqQP36CxHD/view?usp=sharing
> Hello, I noticed that you are using cuml version 0.19 which is more than one year old. Can you try the latest version? We updated TSNE to use `FFT`...
结果：  使用transformers=4.4.2训练了40个epoch结果如下：  两者均未达到您的结果
> The `t` in slerp.py does come from the parameter `t` in the YAML. The way things are set up is that if you set a list of numbers as...