And what you see is that if you train and test in one view, both models do quite well, they get around 80% accuracy.
The Cube2 is first and foremost a music player, something the first-gen model did quite well.
And startups that were better positioned because of lower capital burn, or launched after the lean times with smarter business models, are doing quite well indeed.
Plus-size models can do quite well.
It is worth noting that within the large metropolitan area of London, the Twitter model performs well in absolute terms, and also much better than the radiation model, which does quite poorly in this area (something also found by [29]).
The ANLV model instead does quite a good job at predicting the force, especially during the decay (note that the peak force is somewhat underestimated, though).
But the details of the merger are quite puzzling, as the theoretical models don't quite match what we've observed.
First, it's important to note that there are some set-theoretic concepts which first-order models do capture quite precisely.
All we have out there is models and every time we get data the models don't quite fit what Voyager sees, and then we have to update the models". On Monday this week Voyager rolled 70 degrees anticlockwise as seen from Earth from its normal orientation.
In our case, the generalized QLV model, which belongs to this class of models, does it quite well.
