* Check point callback was initialised with on_best_model callback which overwrote the model on the end of each epoch, even when the valuation loss did not improve
* Fix passing on_best_model callback only if cfg.SEND_BEST_MODEL is set in the config
* Delete model.save() on_best_model as model saving is performed by Keras base class of MyCPCallback.

Add real-time calibration 

2. Add wsCalbirate and calibrate endpoint on web controller

Fix tubhist figure storage location

Fix image in FPS profiling script

Updated Robo HAT MM1 instructions

Websocket client error handling

Added Hackster IO Links
Changed the link for the code.py to internal Donkey Car maintained version.

Update for Robo HAT MM1 code.py

Updated Docs for Simulator (Part 1 - dev)

PR latest

Update comments for serial port.

Add simulator over USB commands

Add websocket to drive api

Allow data augmentation on tub outside of training

Restrict support to python 3.6 and 3.7