Derek Snow

1. I like the roadmap @bin-yang-algotune, I have a view ideas for later down the line: a lot of students come to me for advice on which programs to join,...

That is a pleasure, thanks for your comment.

It's tricky, you can connect it to a Google cloud server. I wouldn't recommend that. I find it easier to just run a Jupyter notebook in the cloud.

Simply put, the training data is never returned to the user the training data is used as both the training and validation set. The training portion is hardcoded at 70%...

> About the model fitting in the training data, Isn't there any value in knowing and returning the perfomance of the models in the training set? I guess it all...

Thanks, seem like there has been a change to an upstream package used by gluonts called pydantic. You can add this to your colab ```!pip install pydantic==1.4```, I have also...

I have made one adjustment, it might possibly work now without an explicit version requirement. If not, I will defer this issue until I redevelop it to be cross-compatible.

That is a cool idea, this package needs so much work!! But it will be interesting if it already performs well.

@rahul4tripathi2 Thanks I always appreciate when someone comes with solutions. Would you be able to submit a pull request when you see changes needed? Best, Derek.

But would this include GBM algorithms like XGBoost and LightGBM, if not, why not, is this technique inherently impossible with these models?