Haven't found anything good yet. I need something similar to Forex, but not Forex: a time serie, numeric data, 5-100 dims, 100k-1M samples.https://twitter.com/fchollet/status/821169875565481984 …
-
if you can find something from sports (player positions, etc), that's pretty good. Or you could use physics engine trajectories.
Thanks for the suggestions, what's "physics engine trajectories"?
-
-
For example, https://www.youtube.com/watch?v=s2DwiwrR1qU … . Put some particles/objects/boids into a physics engine and try to predict a subset of them
-
Upside: infinite data if you want it, downside: kind of far from 'hidden causes' paradigm of most timeseries stuff like HMM
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.