[pdal] Python and chipper filter
Lars
laasunde at hotmail.com
Sun Sep 16 23:32:59 PDT 2018
Hi Adam,
Thanks for a very good suggestion. Will consider changing 😊
A quick comparison showed that they (python and transformation) needed roughly the same time to complete the pipeline.
Regards, Lars
________________________________
Fra: adam steer <adam.d.steer at gmail.com>
Sendt: mandag 10. september 2018 12.45
Til: Lars
Kopi: pdal
Emne: Re: [pdal] Python and chipper filter
Hi Lars
I can’t comment on why your pipeline is slow, but reading the function name of your python script, and picking up on your comment about the python filter, you can achieve (maybe) same result (inverting the Z axis) using:
{
"type":"filters.transformation",
"matrix": "1 0 0 0 0 1 0 0 0 0 -1 0 0 0 0 1"
},
…without going to Python. It’s pretty fast, I did a bunch of processing on the GA MH370 survey with it (4 billion points, very similar - read ASCII, reproject, clean, invert Z, assign all points to ground, write out as LAZ in my case, but could have been pgpointcloud).
So - you could try that and avoid Python in this case.
Also - Intuitively, I’d do the inversion of the Z coordinate before chipping.
Finally, I wonder if there’s a way to test how fast things can be stuffed into your database - to see whether the bottleneck is PDAL or writing to the DB.
Hope that helps, sorry I haven’t directly answered your question, and I’d love to hear anyone else’s thoughts on PDAL processing speed and the python filter.
Adam
On Mon, 10 Sep 2018 at 17:47, Lars <laasunde at hotmail.com<mailto:laasunde at hotmail.com>> wrote:
This appears to be related to the python filter.
Why is the the attached pipeline so slow? Is this a software issue or poorly designed pipeline?
--
Adam Steer
https://www.researchgate.net/profile/Adam_Steer
http://au.linkedin.com/in/adamsteer
http://orcid.org/0000-0003-0046-7236
+61 427 091 712
skype: adam.d.steer
tweet: @adamdsteer
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/pdal/attachments/20180917/9edc27f9/attachment.html>
More information about the pdal
mailing list