[pdal] Python and chipper filter

[adam steer]

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> 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/20180910/bc37b5ea/attachment.html>