[bug #60741] Inaccurate Results for Lambert W function

[Christian Krueger]

Follow-up Comment #1, bug #60741 (group gsl):

Attached is a patch that implements the "Possible Solution" by Daming Zou.
More precisely, gsl_sf_lambert_W0(x) now returns x for |x| < 1e-3 *
GSL_DBL_EPSILON and the first 6 terms of the series expansion from the cited
paper for |x| < 1e-5. Tests are also added.

(file #56747)

    _______________________________________________________

Additional Item Attachment:

File name: bug_60741.patch                Size: 2KiB
    <https://file.savannah.gnu.org/file/bug_60741.patch?file_id=56747>


    AGPL NOTICE

These attachments are served by Savane. You can download the corresponding
source code of Savane at
https://savannah.gnu.org/source/savane-60bab5f4bb97079a4f50fd30c5633b2562628477.tar.gz


    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?60741>

_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/

signature.asc
Description: PGP signature