EeyoreLee

hi @benoitc , I have the same problem. In contrast to this, the object what I wanna shared is a DeepLearning model. it will cost more gpu memory. when I...

> hi @benoitc , I have the same problem. In contrast to this, the object what I wanna shared is a DeepLearning model. it will cost more gpu memory. when...

> @benoitc Thanks for your reply. I use flask warp the deeplearning model to a microservice, then I can use the model by http request. To improve performance and consider...

I just set the environment variable `CXX=g++`. Then it work Updates: Using environment variables is a more reasonable way, otherwise c++ will not be used correctly if using the way...

> 在weibo和resume数据集上按作者提供的代码，没有调参的情况下进行复现。使用的词表为YJ词表。 > resume数据集上的F值为94.45%，加入bert后F值95.73%， > weibo数据集上对应的效果分别为59.71%和66.99%，均达不到原文效果。 > 有复现成功的同学，方便请教一下如果调参的话主要是调整哪些参数吗？ > 另外，数据集是否需要额外的一些处理呢？ > 目前使用的数据集来源： > resume数据集https://github.com/jiesutd/LatticeLSTM > weibo数据集（2nd）https://github.com/hltcoe/golden-horse/tree/master/data weibo的数据我以f=0.685573, pre=0.707379, rec=0.665072复现了。

https://github.com/ear-lee/EarleeNLP/commit/5435e120f8ff13a08b6182ee5e7b750ac030e10d 我用的这个参数，应该就是默认的。run_chinese_ner 直接跑就能复现

> 貌似是我呆瓜了,这01-08每一个都是单独的模型嘛? 加一起是整个模型，你缺少一个pytorch_model.bin.index.json文件，其中保存了模型每个矩阵对应的文件名，你把这个下载一下就可以了

@NostalgiaOfTime @xinghuang2050 - hi, `actor_lora_module_name` is used to filter which linear should be changed to lora linear. And `only_optimize_lora` is used to select whether optimize full-parameters or only the Lora...

@fengy-l @wing7171 - hi, 之前没看到回复，并没有比论文高，论文给出的68.55%，我复现的68.5573%，只是精度舍入的问题

LoRa is not suitable for this situation.