Anshuman Sinha

icon indicating a website link https://anshumansinha16.github.io/

icon company Georgia Institute of Technology icon location Atlanta icon location Student, MS in ML & CSE | Generative AI | Past: Research Engineer @ UIUC | BS in MSE @ IIT Kanpur.

Results 3 comments of Anshuman Sinha

ValueError: The passed save_path is not a valid checkpoint: model/model.ckpt-43000

I am getting this error as well! My backend is : `Using backend: tensorflow.compat.v1'` **code:** ``` # Restore the best test loss model model.restore( save_dir +save_str+"/model.ckpt-" + str(np.argmin(model.losshistory.loss_test)*100), verbose=0) ```...

ValueError: The passed save_path is not a valid checkpoint: model/model.ckpt-43000

> What is your backend? I am using, `Using backend: tensorflow.compat.v1' and getting similar error : [link](https://github.com/lululxvi/deeponet/issues/15#issuecomment-1251218115)

CUDA out of memory : I am using Colab T4 GPU

No, I have not yet solved this issue.