652994331
652994331
@brightmart 是因为我使用您的项目中的run_classifier 可以使用相对较大的batch size ,似乎batch 的提高对效果有所帮助,是这样的原因
@feifeibear 不好意思 这图是之前改了忘改回来了 我用的是1。 还是存在相同的问题, 我理解这个脚本必须是 tf2.0的模型才可以使用的对吧 huggingface 的bert 我试了一个convert_huggingface_bert_to_npz bert-base-uncased bert_tf.npz 是可以的
@artmatsak Hi, i am trying to do the same thing, have you figure it out? thanks for your reply
您好 我现在自己产生的预训练数据集还要些问题,可以请教下您吗,谢谢。详细问题在这里https://github.com/lonePatient/electra_pytorch/issues/9 , 应该是使用流程上的问题
@MaartenGr thanks for your help, below are my training code and looks like the version of my bertopic is 0.11.0 and my cuda is 10.2, torch version is 1.12.0+cu10.2 e
@MaartenGr sure, the codes are as below:
@MaartenGr thank you so much. I found something this afternoon, it seems my cuda version is 10.2 and it requires 11.2 which includes lib_cudart.so.11. right after I installed cuda11.2, the...
@yemanzhongting 嗯, 那是我之前写的,当时搞那个搞一半搞不下去在气头上,是我暴躁了,我道歉。 PS:后来这个事儿我解决了,综合了网上的各种帖子。我只能说写GIT 很多人确实是写的比较随意,放上来再就不管了,对事不对人
@yywhsgnd hi, I met the same question, have u solved it ?
@ I see, so grid-search is a tool to find all the parameters setting and you actually need to evaluate which setting is the best. the way you do the...