[FEATURE] Enhancing Train Script Functionality with Tensorboard Integration

**Is your feature request related to a problem? Please describe.**
I would greatly appreciate it if a flag could be implemented, allowing me to log accuracy in Tensorboard format after each epoch.
**Describe the solution you'd like**
I kindly request the addition of a flag, allowing for the logging of accuracy to Tensorboard format after each epoch
**Describe alternatives you've considered**
Regarding alternative methods, I have contemplated saving the output to a text file for manual log inspection, as well as utilizing the functionality of Excel to effectively compare graphs.
**Additional context**

Start by locating the repository's train script and its existing epoch-level metric logging. Check how command-line flags are defined and how training accuracy is calculated. Done means a flag enables accuracy logging in TensorBoard format after each epoch, with the resulting logs usable for graph comparison.

Written by the indexing model from the issue text.