Ramlinbird

> > Okay, so I've plotted the teacher and student variances from two test runs with a small dataset and only for a few epochs: > > https://imgur.com/a/dynF7P9 > >...

> > > > > > > > > > Okay, so I've plotted the teacher and student variances from two test runs with a small dataset and only for...

@alexeib Thanks a lot for your quick reply! According to your loss plot, my training seems all right? And I also check my variance plot,  However, the pred var...

> > > ga.Chrom = np.random.randint(0,2,size=(80,20)) > > > 这里的80和20，分别表示什么意思？ > > > > > > 种群规模和基因长度，这个大小和初始化后的 ga.Chrom 保持一致即可 > > 能否添加 `ga.x2chrom()` 功能？直接指定二进制的初始种群基因还是十分不便 是的，如果有x2chrom功能就更好了😂

> > > > > ga.Chrom = np.random.randint(0,2,size=(80,20)) > > > > > 这里的80和20，分别表示什么意思？ > > > > > > > > > > > > 种群规模和基因长度，这个大小和初始化后的 ga.Chrom 保持一致即可 >...

> @WangHexie 所有词向量文件里面的token都是按字符频率排序的，从高到低依次减少。 现在想基于词向量，生成句向量，想要参考SIF的做法，里面会有用到词频文件。可否将词频也分享一下呢～（百度百科语料）

> Hello, sure we can！ we support the pre-training of ERNIE in following folder. > > https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/ernie-1.0 > > you can try it！ Thanks for your quick reply. I don't...

Thanks a lot.

@Spijkervet Hello, sorry to disturb. I tried change the audio_length, and it failed. I read the source code and found that the encoder of audio is sample_cnn which can't process...

> @Spijkervet Looking for your reply! Thanks a lot I met this error also, just change the target datatype to long inside the accuracy calcation logic works.