Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

3.0bata版本的lstm参数输出有误 #70245

Open
sealoongleft opened this issue Dec 16, 2024 · 3 comments
Open

3.0bata版本的lstm参数输出有误 #70245

sealoongleft opened this issue Dec 16, 2024 · 3 comments
Assignees

Comments

@sealoongleft
Copy link

sealoongleft commented Dec 16, 2024

bug描述 Describe the Bug

问题1:lstm输出参数时weight_ih_l1的形状为[3,2],bias_ih_l1的形状为[12,2],这明显不对啊,偏置元素比weight还多。
问题2:lstm的1.cell.weight_ho权重无法访问,我用的是lstm.1.cell.weight_ho访问的,但是报错了。用这个方法访问别的lstm权重偏置就可以唯独1.cell.weight_ho这个不行。
问题3:1.cell.weight_ho中的1和ho代表什么意思。没想明白。

复现代码如下:

import paddle.nn as nn

lstm=nn.LSTM(input_size=4,hidden_size=3,num_layers=2,proj_size=2)
for a in lstm.named_parameters():
    print(a)

其他补充信息 Additional Supplementary Information

win11
其他用的都是当前paddle最新gpu版本
api版本是:3.0 beta

补充:如果可以的话,能不能告知下大概多久能修复此问题。论文中使用了这个api,拖太久的话我就删掉这个proj_size参数了。感谢技术大佬。

@sealoongleft
Copy link
Author

anybody here?

@zoooo0820
Copy link
Contributor

你好,问题已经反馈给该API负责同学排查了,有结论后尽快同步在这里呢

@sealoongleft
Copy link
Author

sealoongleft commented Dec 18, 2024 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants