Hi, > On May 17, 2026, at 17:16, Dean Rasheed <[email protected]> wrote: > > OK, here's a more complete patch along those lines, intended to apply > on top of the regr_r2() patch.
Thanks for the regr_intercept.patch. The approach looks good to me. I only noticed a few small things: 1. The patch file seems to have a format issue and doesn't apply directly. `git apply` reports: ``` error: git apply: bad git-diff - expected /dev/null on line 2 ``` 2. `dy` seems a bit hard to understand. Perhaps `offset`, as used in the earlier sketch, would be clearer? 3. Do we need to add tests for the underflow path, and perhaps for the Inf/NaN guard? Other than that, this looks good to me. -- Best regards, Chengpeng Yan
