+1 There's still more digging for me to do on some of the content diffs, but the few Japanese docs I manually reviewed looked better in 2.0.21 than they did before...I suspect something is going wrong with tika-eval. Side note, I opened a ticket to look into all the Nepali detections, which have also appeared in some corpora in my $dayjob.
In short, looks good! Thank you! On Thu, Aug 13, 2020 at 11:27 PM Tilman Hausherr <[email protected]> wrote: > +1 > > > Tilman > > Am 13.08.2020 um 21:44 schrieb Andreas Lehmkuehler: > > Am 10.07.20 um 08:25 schrieb Andreas Lehmkuehler: > >> Hi, > >> > >> how about cutting a 2.0.21 release in about 2 weeks from now? The > >> current release doesn't seem to be that good and I'd like to get out > >> a new one asap. > > The latest test results are looking good. I'm planning to cut the > > release next Monday if nobody objects > > > > Andreas > > > >> > >> WDYT? > >> > >> Andreas > >> > >> --------------------------------------------------------------------- > >> To unsubscribe, e-mail: [email protected] > >> For additional commands, e-mail: [email protected] > >> > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [email protected] > > For additional commands, e-mail: [email protected] > > > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] > >
