aclements / go-misc Goto Github PK

Hi,

I've implemented a tool to fit benchmarks to least squares models at benchls. I didn't know about this repo at the time. It seems like there is a (currently unused?) LinearLeastSquares model in this repo, which I assume is intended for use with the plotting functionality, or perhaps for reporting?

Would you like for me to add code to calculate the confidence interval to benchmany? Or for that matter any of its functionality. I'm trying to get a good benchmarking tool for visualizing performance for gonum, and it looks like you're doing better than anything I have.

This would be usefull to easily see how a certain benchmark changes over time, to identify commits that regressed and improved performance of this benchmark.