guangyliu/latentops

Source code of LatentOps

𝔾𝕦𝕒𝕟𝕘𝕪𝕚 𝕃𝕚𝕦 𝒫𝑜𝓈𝓉𝒹𝑜𝒸 @ 𝑀𝐵𝒵𝒜𝒰𝐼

latentops's Issues

修改输出文件目录

你好！我尝试修改conditional_generation.sh中输出文件的目录，但是在我的新文件夹中并没有文件输出，请问有什么可能的原因嘛？

Add more instructions to train vae/classifer/gan?

Hi, @guangyliu , that's really an interesting and helpful work.

Recently, I'm trying to conduct experiments on my own dataset(agnews), but I find it difficult for me to run your code. So it would be great if you can add more docs about how to train the vae/classifer/gan.

Here are something I think might be important and useful:

1. Dataset preparation, including data format and detailed hyperparameters.
2. Traing scripts. For example, I tried to train my own latent classifer using the script code/train_classifier_latent.sh, but I'm not sure whether it is correct. Furthermore, I cannot find any scripts to train the gan. Maybe I missed something important.

Thanks in advance! Looking forward to your reply.

在大模型上有没有什么样的办法同样加入隐向量？

看到了你们的工作真的给了我很大的启发，我现在想结合你们的代码来做大语言模型的尝试。
VAE的decoder我选择了现在比较实用的chatglm-6B，但是训练的过程中发现loss虽然在下降，但是eval_bleu却始终是0，不知道你们是否在大语言模型中有过类似的尝试，希望可以得到一些经验，谢谢。

Where is conditional_generation.sh?

How to achieve text editing?

Hello! Thanks for your nice job! As shown in your paper, the performances on text editing with single or compositional attribute are quite impressive. I wonder how to achieve this purpose based on your codes and checkpoints? I cannot find any instruction on this in your repo. Looking forward to your reply!

condition_genaration.py中数据是如何读取的？

你好，我想问下condition_genaration.py中数据是如何从给定的文件被读取的，我似乎没有找到.

論文中attributes with keywords

不好意思，我想請問一下，論文中generation with compositional attributes with keywords這部分的code放在github中的哪裡？
非常感謝作者您百忙之中還願意查看我的問題，期待您的回覆~

A BUG

We followed your Readme.md, but I found this problem:

11/01/2022 21:15:14 - INFO - main - Pre-trained Optimus is successfully loaded
Traceback (most recent call last):
File "examples/big_ae/conditional_generation.py", line 511, in
main()
File "examples/big_ae/conditional_generation.py", line 458, in main
model_vae.to(args.device) #
File "/home/ubuntu/anaconda3/envs/RL/lib/python3.8/site-packages/torch/nn/modules/module.py", line 927, in to
return self._apply(convert)
File "/home/ubuntu/anaconda3/envs/RL/lib/python3.8/site-packages/torch/nn/modules/module.py", line 579, in _apply
module._apply(fn)
File "/home/ubuntu/anaconda3/envs/RL/lib/python3.8/site-packages/torch/nn/modules/module.py", line 579, in _apply
module._apply(fn)
File "/home/ubuntu/anaconda3/envs/RL/lib/python3.8/site-packages/torch/nn/modules/module.py", line 579, in _apply
module._apply(fn)
[Previous line repeated 3 more times]
File "/home/ubuntu/anaconda3/envs/RL/lib/python3.8/site-packages/torch/nn/modules/module.py", line 602, in _apply
param_applied = fn(param)
File "/home/ubuntu/anaconda3/envs/RL/lib/python3.8/site-packages/torch/nn/modules/module.py", line 925, in convert
return t.to(device, dtype if t.is_floating_point() or t.is_complex() else None, non_blocking)
RuntimeError: CUDA error: unknown error

Do you know how to fix it ???