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

How to reload the model in a different machine with same hardware #10

Open
DebayanChakraborty opened this issue Aug 14, 2020 · 1 comment
Assignees
Labels
Inference Issue related to inference

Comments

@DebayanChakraborty
Copy link

Hi ,

Firstly great tutorial . The multiclass model tutorial really helped me in understanding how to leverage BERT . I have my model saved as shown post training. However when i am loading on another machine i see drop in accuracy . At the same time, it does not always work .

Could you guide with a proper code snippet to load the models .

I do have the ability to define the class in the new environment so that is not an issue

@abhimishra91
Copy link
Owner

abhimishra91 commented Aug 17, 2020

Hi @DebayanChakraborty , thank you for the feedback and using the tutorials.

What is change in performance? If its marginal, then that is pausible due to different seed values for various packages. The performance should not be changing drastically. In cases where its not working, do you have any error log that you can share?

In terms of how to load the models, you can check how i have used the models in my other project. Project Insight.

Have a look at any of the backend services code in the src_fastapi directory. I am leveraging the models trained with these notebooks in that project and have not faced any issues yet. So keen to see what error you are facing.

@abhimishra91 abhimishra91 self-assigned this Aug 17, 2020
@abhimishra91 abhimishra91 added the Inference Issue related to inference label Aug 17, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Inference Issue related to inference
Projects
None yet
Development

No branches or pull requests

2 participants