Is there any dev roadmap for constructing/visualising dynamic computational graphs (defined through eager execution)?
Hi @bairesearch,
Apologize for the delay in response.
tf.function compiles Python functions into optimized computational graphs, enabling efficient execution while maintaining eager execution mode. It transforms the function into a TensorFlow graph that can be executed faster, leveraging hardware accelerators and distributed resources. tf.function can be applied either as a decorator or via direct calls.For more details, please refer to the documentation on graphs and improving performance with tf.function.
Thank You.