This .py file has a class "TBSentenceTransformer" that is inherited from SentenceTransformer. This class is to be used instead of SentenceTransformer if we want to log our training progress using TensorBoard. 
I have been waiting for the SentenceTransformer team to integrate tensorboard into their repo but it seems like they are taking some time doing it since it is ground up. This is an alternative to that. Code has been written to minimize the impact on any of the repos using SentenceTransformers i.e. if at any time, SentenceTransformers do come up with a reporting framework, it will be easy to revert to that if required with minimal changes to the code

A couple of lines added to the creation of the SentenceTransformersFinetuneEngine class:
1. Add a parameter log_path that specifies where the tensorboard logs should be stored
2. Update the constructor: self.model = = TBSentenceTransformer() if the log_path is specified else self.model = SentenceTransformer()

Updated class inherited from SentenceTransformers to include tensorboard reporting

Contains an inherited class from SentenceTransformer called TBSentenceTransformer that allows llamaindex's SentenceTransformersFinetuneEngine to track and record loss while finetuning using tensorboard.
Credit to this PR: UKPLab/sentence-transformers#1532 , by Ahmed Magdy https://github.com/Mogady

Updated SentenceTransformersFinetuneEngine's self.model to accept a TBSentenceTransformer object if a log_path is specified or revert to accepting a SentenceTransformer object. Initial commits had some mistakes..