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

Dimension mismatch Error #1801

Open
SSwethaSel0609 opened this issue Nov 15, 2024 · 2 comments
Open

Dimension mismatch Error #1801

SSwethaSel0609 opened this issue Nov 15, 2024 · 2 comments

Comments

@SSwethaSel0609
Copy link

I have downloaded the pretrained model from icefall. I have tried to finetune the model in zipformer using my own data. It is showing mismatch error. I have attached the error file
error.txt

@csukuangfj
Copy link
Collaborator

From your error log:

        size mismatch for encoder.encoders.0.layers.0.self_attn_weights.in_proj.weight: copying a param with shape torch.Size([272, 192]) from checkpoint, the shape in current model is torch.Size([544, 192]).

The first thing you have to do is to find out the exact parameters used to train the model that you are using for finetuing.

@SSwethaSel0609
Copy link
Author

I solved that error. Can you please tell me this isse?
File "finetune.py", line 1121, in train_one_epoch
cur_grad_scale = scaler._scale.item()
AttributeError: 'GradScaler' object has no attribute '_scale'

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

No branches or pull requests

2 participants