Jack Poulson
Jack Poulson
There is a long history of proposals to include simple modifications of `?syrk` and `?herk` to support `C += alpha A (D A)^T`, where D is diagonal. This kernel turns...
BLIS often refers to fused level 1 BLAS-like operations, but I have not seen any fused level 2 operations (e.g., a single-sweep y := A x and u := A^T...
It seems that the URLs have moved from `http://www.nacis.org` to `http://www.naciscdn.org`, so I have updated the `README.md` accordingly.
After launching a server from the current source (and using the updated data download scripts described in https://github.com/cambecc/earth/pull/97), I can manipulate the underlying spherical earth model, but I get the...
The documentation for the deflation process in {s,d}lasd2 claims that the deflated singular values are sorted in _increasing_ order, but they are in fact more-or-less sorted in _decreasing_ order, with...
The "--output" and "--model" arguments to onmt_release_model were incorrectly specified with single hyphens, which results in "unrecognized argument" errors if run verbatim.
It appears that using `Census.ALL` instead of something of the form `{'for': 'state:*'}` does not work, even on the core example in the README: ``` >>> c.acs5.get('B01001_004E', Census.ALL) Traceback (most...
It appears that sigma always assumes double-precision real data. Are there plans to support, for example, complex data? (Support for double-double and/or quad-precision would also be very useful.)
There does not currently seem to be a trivial way to make a deep copy of a graph (e.g., with `copy`).
It seems to me that Eq. 8 from [1] is not necessarily satisfied when the input matrix for LLL has complex values. Indeed, said paper seems to embed complex matrices...