Is it possible to use lr_fund if the model is not build by fastai

If I use Keras to build a model, can I still use lr_fund to find a learning rate and also cyclic learning rate function of fastai?

The short answer is No. But you can take a look at the code under the hood and re-implement it to work with another DL framework if that’s your use case.

1 Like

Thank you very much, Ramesh!

1 Like