[GRASS-web] i.cluster module problem

[Moritz Lennert]

[Please always keep discussions on the mailing list]

On 15/02/18 15:27, Marta Fernandez wrote:
> Thank you for respond so fast.
> 
> Input data: (1 from 7)
> 
> | Map:      composite.1 at Marta              Date: Thu Feb 15 09:31:14 
> 2018    |
> 
> | Mapset:   Marta                          Login of Creator: 
> Marta           |
> 
> | Location: 
> prueba                                                           |
> 
> | DataBase: 
> C:\Users\Marta\grass                                             |
> 
> | Title:    
> composite.1                                                      |
> 
> | Timestamp: 
> none                                                            |
> 
> |----------------------------------------------------------------------------|
> 
> |                                                                            
> |
> 
> |   Type of Map:  raster               Number of Categories: 
> 0               |
> 
> |   Data Type:    
> FCELL                                                      |
> 
> |   Rows:         
> 6686                                                       |
> 
> |   Columns:      11422                        
>                                |
> 
> |   Total Cells:  
> 76367492                                                   |
> 
> |        Projection: UTM (zone 
> 29)                                           |
> 
> |            N: 4725034.75771    S: 4724499.87771   Res:  0.08 
>                |
> 
> |            E: 553994.32785    W: 553080.56785   Res:  
> 0.08                 |
> 
> |   Range of data:    min = 0.06990715  max = 
> 0.6211307                      |
> 
> |                                                                            
> |
> 
> |   Data 
> Description:                                                        |
> 
> |    generado por 
> r.in.gdal                                                  |
> 
> |                                                                            
> |
> 
> |   Comments:  
>                                                                |
> 
> |    r.in.gdal -k 
> input="A:\Marta\Documents\uforest\pruebas\pruebas_clasi\   |
> 
> |    ficacion_habitats\clasificacion_nosupervisada_grass\composite.tif" 
> o\   |
> 
> |    utput="composite" memory=300 offset=0 
> num_digits=0                      |
> 
> |                                                                            
> |
> 
> +----------------------------------------------------------------------------+
> 
> Region data:
> 
> (Thu Feb 15 15:25:37 2018)
> 
> g.region -p
> 
> projection: 1 (UTM)
> 
> zone:       29
> 
> datum:      wgs84
> 
> ellipsoid:  wgs84
> 
> north:      4725034.75771
> 
> south:      4724499.79771
> 
> west:       553080.56785
> 
> east:       553994.32785
> 
> nsres:      0.08
> 
> ewres:      0.08
> 
> rows:       6687
> 
> cols:       11422
> 
> cells:      76378914
> 
> (Thu Feb 15 15:25:37 2018) Comando finalizado (0 segundos)
> 
> Marta Fernández Álvarez
> Grupo de Investigación CI5 – Lab. Ingeniería Cartográfica
> EI Forestal-Universidad de Vigo
> 
> *De: *Moritz Lennert <mailto:mlennert at club.worldonline.be>
> *Enviado: *jueves, 15 de febrero de 2018 14:58
> *Para: *Marta Fernandez <mailto:martaferal11 at gmail.com>
> *CC: *grass-web at lists.osgeo.org <mailto:grass-web at lists.osgeo.org>
> *Asunto: *Re: [GRASS-web] i.cluster module problem
> 
> Am 15.02.2018 12:43 schrieb Marta Fernandez:
> 
>  > Good morning,
> 
>  >
> 
>  > I´m Marta, researcher from Vigo University. We were trying classify a
> 
>  > band set (7 bands), so we introduced it in i.cluster module and leave
> 
>  > the configuration options by default. The console returns the
> 
>  > following error "Number of insufficient sample points, please run
> 
>  > again and choose a larger sample size." We try to change the size of
> 
>  > the sampling interval, but it returns an error “ERROR: Value (s) of
> 
>  > sampling intervals (50,50) not supported". we have also tried to
> 
>  > express these values in another way, and testing with values of
> 
>  > greater and lesser magnitude.
> 
>  >
> 
>  > Could you tell me what the error is?
> 
> What are your region settings ?
> 
> g.region -p
> 
> compared to your input data:
> 
> r.info OneInputBand ?
> 

Ok. Region does not seem to be the issue.

Can you give us the precise manner in which you called i.cluster ? And 
the content of the group you use ? This might also help understand where 
the error after giving a sample size comes from.

Moritz