Hi Team I’m trying to deploy a tensorflow model that’s a part of TFRS library, this model is built and trained in a way that cannot be saved in a single file using the SavedModel format, but only can be saved using manual saving of weights

# Save the weights
model.save_weights('./checkpoints/my_checkpoint')

# Create a new model instance
model = create_model()

# Restore the weights
model.load_weights('./checkpoints/my_checkpoint')

# Evaluate the model
loss, acc = model.evaluate(test_images, test_labels, verbose=2)
print("Restored model, accuracy: {:5.2f}%".format(100 * acc))

This isn’t supported using the current

bentoml.tensorflow.save_model

or

bentoml.keras.save_model

, is there any workaround to have a custom load or save to create a runner out of it?