Git Gao

I'm creating a simple HMM with gaussian emission. Modified from example from: https://pyro.ai/examples/hmm.html I got two issues: 1. sample function always return [0, 1, 2] after first epoch. 2. parameter...

tqdm progress bars in colab, jupyterlab

18

The newest version isn't showing the progress of each chain as before. I tried to do show_progress = 1 and doesn't work. How do I make it show the progress...

I'm trying to calculate the RMSE to do dimensionality selection but the model only gives me the matrix with data imputed. Is there any way to get the inverse transformed...

[Bug]: Chroma.add terminates flask without any error

2

### What happened? I'm creating an API with Flask. The other side will send me a file and I will save it to chroma database on my side. Chroma.add will...

对话数据怎么设置不对别人说的话训练？

3

通常对话数据只在模型的回复上面进行梯度下降。比如把非模型输出的label index变成-100。 类似这样： ``` user 你好 assistant 请问有什么帮助你的？ ``` transformer只在assistant后面进行梯度下降。 rwkv这种结构是不是不太适合这种只在回复上面训练的方式？如果适合，请问在RWKV-LM训练的时候是怎么设定的？