After having succeeded in compiling i.pr from the AddOns svn (thanks to Glynn), I am now trying to get it to work in grass6.3. I already contacted the original author, but he hasn't used it for a long time and cannot help.
Documentation is a bit scarce, but from what I gather i.pr.training
is the starting point. However, trying to work with it I only get a fairly unusable screen (see screenshot at [1]). I launched the module with:
i.pr.training map=`g.mlist type=rast pattern="spot.ms*" sep=","`
training=ipr_training4 rows=3 cols=3 class=1 vis_map=spot.comp
Could someone tell me if I am doing anything wrong ?
Moritz
[1] http://geog-pc40.ulb.ac.be/grass/misc/
Hi Mortiz,
I looked at i.pr a while ago, but didn't try
the interactive mode for point selection.
If the problem you are encountering persists,
you could digitise a vector map of labeled training
locations, which you could then use in i.pr.training.
This will bypass the need to interactively select training
points when you run the module.
i.pr.training map=`g.mlist type=rast pattern="spot.ms*" sep=","`
training=ipr_training4 vector=YourLabeledVector rows=3 cols=3
vis_map=spot.comp
Hope this helps. Would be interested in hearing how
you get on.
Regards,
Daniel.
Moritz Lennert wrote:
After having succeeded in compiling i.pr from the AddOns svn (thanks to Glynn), I am now trying to get it to work in grass6.3. I already contacted the original author, but he hasn't used it for a long time and cannot help.
Documentation is a bit scarce, but from what I gather i.pr.training
is the starting point. However, trying to work with it I only get a fairly unusable screen (see screenshot at [1]). I launched the module with:
i.pr.training map=`g.mlist type=rast pattern="spot.ms*" sep=","`
training=ipr_training4 rows=3 cols=3 class=1 vis_map=spot.comp
Could someone tell me if I am doing anything wrong ?
Moritz
[1] http://geog-pc40.ulb.ac.be/grass/misc/
_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user
Hi Daniel,
On 28/11/07 13:12, Daniel McInerney wrote:
Hi Mortiz,
I looked at i.pr a while ago, but didn't try
the interactive mode for point selection.
If the problem you are encountering persists,
you could digitise a vector map of labeled training
locations, which you could then use in i.pr.training.
This will bypass the need to interactively select training
points when you run the module.
i.pr.training map=`g.mlist type=rast pattern="spot.ms*" sep=","`
training=ipr_training4 vector=YourLabeledVector rows=3 cols=3
vis_map=spot.comp
Thanks for the tip. Tried it, but doesn't seem to work. Here is what I get, i.e. still a black screen:
http://geog-pc40.ulb.ac.be/grass/misc/i_pr2.png
And the resulting ipr_training4 contains only the header:
Data type:
GrassTraining
Number of layers:
3
Label:
ipr_training4
Data:
Layer_1 Layer_2 Layer_3 Class East North Rows Cols EW-res NS-res
Moritz
Hi Moritz,
What does v.info training give? Did
you use v.digit or v.in.ascii to
create the layer? I used the latter
when I was working with it.
Regards,
Daniel.
Moritz Lennert wrote:
Hi Daniel,
On 28/11/07 13:12, Daniel McInerney wrote:
Hi Mortiz,
I looked at i.pr a while ago, but didn't try
the interactive mode for point selection.
If the problem you are encountering persists,
you could digitise a vector map of labeled training
locations, which you could then use in i.pr.training.
This will bypass the need to interactively select training
points when you run the module.
i.pr.training map=`g.mlist type=rast pattern="spot.ms*" sep=","`
training=ipr_training4 vector=YourLabeledVector rows=3 cols=3
vis_map=spot.comp
Thanks for the tip. Tried it, but doesn't seem to work. Here is what I get, i.e. still a black screen:
http://geog-pc40.ulb.ac.be/grass/misc/i_pr2.png
And the resulting ipr_training4 contains only the header:
Data type:
GrassTraining
Number of layers:
3
Label:
ipr_training4
Data:
Layer_1 Layer_2 Layer_3 Class East North Rows Cols EW-res NS-res
Moritz
On 28/11/07 14:25, Daniel McInerney wrote:
Hi Moritz,
What does v.info training give?
> v.info training
+----------------------------------------------------------------------------+
| Layer: training |
| Mapset: user1 |
| Location: imagery60 |
| Database: /home/mlennert/GRASS/DATA |
| Title: |
| Map scale: 1:1 |
| Map format: native |
| Name of creator: mlennert |
| Organization: |
| Source date: Thu Nov 22 16:22:31 2007 |
|----------------------------------------------------------------------------|
| Type of Map: vector (level: 2) |
| |
| Number of points: 0 Number of areas: 14 |
| Number of lines: 0 Number of islands: 14 |
| Number of boundaries: 14 Number of faces: 0 |
| Number of centroids: 14 Number of kernels: 0 |
| |
| Map is 3D: 0 |
| Number of dblinks: 1 |
| |
| Projection: Projection inconnue |
| N: -1178.27114746 S: -2050.34100516 |
| E: 1841.24981384 W: 780.66696671 |
| |
| Digitization threshold: 0 |
| Comments: |
| |
+----------------------------------------------------------------------------+
Did
you use v.digit or v.in.ascii to
create the layer? I used the latter
when I was working with it.
I used v.digit.
Moritz
I'm under the impression that i.pr.training
requires a point vector map, but I'm not 100%
sure and haven't tested a polygon vector file?
Perhaps, you could use the polygon centroids
to extract your training features or else
digitise a point vector map.
Regards,
Daniel.
Moritz Lennert wrote:
On 28/11/07 14:25, Daniel McInerney wrote:
Hi Moritz,
What does v.info training give?
> v.info training
+----------------------------------------------------------------------------+
| Layer: training |
| Mapset: user1 |
| Location: imagery60 |
| Database: /home/mlennert/GRASS/DATA |
| Title: |
| Map scale: 1:1 |
| Map format: native |
| Name of creator: mlennert |
| Organization: |
| Source date: Thu Nov 22 16:22:31 2007 |
|----------------------------------------------------------------------------|
| Type of Map: vector (level: 2) |
| |
| Number of points: 0 Number of areas: 14 |
| Number of lines: 0 Number of islands: 14 |
| Number of boundaries: 14 Number of faces: 0 |
| Number of centroids: 14 Number of kernels: 0 |
| |
| Map is 3D: 0 |
| Number of dblinks: 1 |
| |
| Projection: Projection inconnue |
| N: -1178.27114746 S: -2050.34100516 |
| E: 1841.24981384 W: 780.66696671 |
| |
| Digitization threshold: 0 |
| Comments: |
| |
+----------------------------------------------------------------------------+
Did
you use v.digit or v.in.ascii to
create the layer? I used the latter
when I was working with it.
I used v.digit.
Moritz
I have added a funny machine translation to English
of the usage.tex file into
grassaddons/i.pr/doc/usage_EN.txt
this could be the start for a description.html file...
Markus
Daniel McInerney wrote:
I'm under the impression that i.pr.training
requires a point vector map, but I'm not 100%
sure and haven't tested a polygon vector file?
Perhaps, you could use the polygon centroids
to extract your training features or else
digitise a point vector map.
Regards,
Daniel.
--
View this message in context: http://www.nabble.com/i.pr%3A-screen-does-not-show-map-tf4887766.html#a14108977
Sent from the Grass - Users mailing list archive at Nabble.com.
