Keras just for load_model

[marcelo-bn]

Hey guys!

I only need to load a model into a production environment, but to do this I need to install the entire Keras library and consequently the entire tensorflow library. All these packages cost me almost 1.5 GB and my resources are limited.

So I need a simple version of Keras just to load my model. Is there any way to do this?

[sachinprasadhs]

If you do not have any tf.data dependency on your model, you can set pytorch as a backend with the Keras 3. PyTorch takes the less installation memory across all 3 backends(JAX, TensorFlow, PyTorch) which Keras supports.

[github-actions[bot]]

This issue is stale because it has been open for 14 days with no activity. It will be closed if no further activity occurs. Thank you.

[lxzheng]

@marcelo-bn You can try tf serving. It is designed specifically for serving models in production.

[github-actions[bot]]

This issue is stale because it has been open for 14 days with no activity. It will be closed if no further activity occurs. Thank you.

[github-actions[bot]]

This issue was closed because it has been inactive for 28 days. Please reopen if you'd like to work on this further.

[google-ml-butler[bot]]

Are you satisfied with the resolution of your issue? Yes No