i want to add macro average of F1Score to model metrics in tensorflow and watch it at each epoch

I want to use multiple GPUs to train a model, because there is not enough memory in a single GPU card for my training.

Error description:

In TensorFlow’s offcial documentations, they always pass training=True when calling a Keras model in a training loop, for example, logits = mnist_model(images, training=True).

In TensorFlow’s offcial documentations, they always pass training=True when calling a Keras model in a training loop, for example, logits = mnist_model(images, training=True).

In TensorFlow’s offcial documentations, they always pass training=True when calling a Keras model in a training loop, for example, logits = mnist_model(images, training=True).

In TensorFlow’s offcial documentations, they always pass training=True when calling a Keras model in a training loop, for example, logits = mnist_model(images, training=True).

I use GPU to train the model on Kaggle while the code and dataset from the book Deep Learning With Python second edition.
the kaggle’s tensorflow version is 2.15.0 and keras’version is 3.4.1.

I am trying to run a GradCAM that gives me the below error. I am confused as to why this is as Sequential is clearly being called. I have tried uninstalling tensorflow, restarting the kernel, and trying other environments. Any help is appreciated.

I am a newbie learning keras following this tutorial