Hello 
I’ve been working on ways to improve the collaborative training, transparency & reproducibility for open-source ML with Keras 
With the recent features in Hugging Face Hub, you can simply push your models to Hugging Face and let others try it with only one line of code 
This will push your TensorBoard traces and host them open to everyone, generate a model card that includes your model’s hyperparameters, metrics and architecture and host your model itself so anyone can load it with another line of code! 
I’ve written a blog post about this, would love to hear your opinion 
Also see a hosted example model with model card and tensorboard traces in this link.
