meta / prophet
Tool for producing high quality forecasts for time series data that has multiple seasonality with linear or non-linear growth.
100out of 100
The new stars are spread over many days, and the repo is forked at a rate that fits its stars.
| Stars on a few days | 11% of the 106 new stars in 60 days came on the 3 biggest days | −0 of 30 |
|---|---|---|
| Forks for the stars | 227 forks for every 1,000 stars | −0 of 50 |
An estimate from the star counts and fork count of Oct 30, nothing else. It describes the shape of the numbers and says nothing about why they look that way: a launch, a post that went viral or a link from a big account give the same shape. The rule and how often it was right.
Total stars
All stars added up, from GitHub's public star counts
+56 stars in 30 days, Sep 30 to Oct 30
Latest day, Oct 30: 20,464 stars, +1
Dashed line, up to Oct 23: counted back from our first count with GitHub's new stars per day (stars that people took back are not in them, so it can run a little off). Solid line after that: our own daily star counts.