TensorBoard

TensorBoard is one of the most important strengths of the TensorFlow platform and with TF 2.0, TensorBoard has gone to the next level. In machine learning, to improve your model weights, you often need to be able to measure them. TensorBoard is a tool for providing the measurements and visualizations needed during the machine learning workflow. It enables tracking experiment metrics such as loss and accuracy, visualizing the model graph, projecting embeddings to a lower dimensional space, and much more. In contrast to TF 1.x, TF 2.0 provides a very simple way to integrate and invoke TensorBoard using callbacks, which were explained in The fit() API section. Also, TensorBoard provides several tricks to measure and visualize your data and model graphs, and it has a what-if and profiling tool. It also extends itself to be able to debug.

...