XVALID Process
To access this process:
- Enter "XVALID" into the Command Line and press ENTER.
-
Display the Find Command screen, locate XVALID and click Run.
See this process in the Command Table.
Process Overview
Note: This is a superprocess and running it may have an effect on other Datamine files in the project.
XVALID assists in the selection of parameters for grade estimation, using the cross-validation method.
The input data file is the sample data which will later be used for estimating grades into a block model. For kriging it allows different model variograms to be tested and compared. For inverse power of distance it allows different powers to be compared.
The input to XVALID is consistent with the input required for the grade estimation process ESTIMA. In fact the three input parameter files are identical for the two processes.
The cross-validation method works by removing each point in turn from the data file and estimating its value from the remaining data. In this way a table of actual and estimated values is created. A detailed statistical analysis is then carried out comparing the actuals and estimates. One or more of the estimation parameters can then be changed and the process rerun to see whether the new parameters improve the results of the statistical analysis. The method is therefore iterative, requiring several runs to establish the best set of parameters.
XVALID uses the block model grade estimation process ESTIMA to do the actual interpolation. Therefore part of the text on the Output window is the same as for ESTIMA, and will refer to a block model rather than points. However the number of discretisation points is set to 1x1x1 so in effect point estimation is being used.
There are three input parameter files all of which can be set up using the ESTIMATE process or the Table Editor:
-
SRCPARM: Search Volume Parameter File
-
ESTPARM: Estimation Parameter File
-
VMODPARM: Variogram Model Parameter File
and two other input files:
-
IN: Sample File
-
VGRAM: Experimental Variogram File
The results of the cross-validation are written to three separate output files, all of which are optional:
-
XVSAMPS: cross-validated output sample file
-
XVSTATS: cross-validation statistics
-
SAMPOUT: sample output file containing weights for each estimate
The statistics are also displayed in the Output window as illustrated in the example. When the first set of results have been calculated and displayed you are then able to select from a menu which allows you to edit one or more of the parameter files or examine the results, and then rerun the cross-validation:
1 - edit search volume parameter file spar10
2 - edit estimation parameter file epar10
3 - edit variogram model parameter file vpar10
4 - examine variogram model interactively (VARFIT)
5 - examine cross-validation statistics file xxstats
6 - delete cross-validation statistics file xxstats
7 - create plot of actual v estimate
8 - re-run cross-validation
0 - exit cross-validation
Option 4 allows you to use the interactive variogram fitting process VARFIT. In order to use this option you must have specified the experimental variogram file VGRAM when selecting XVALID.
Options 5 and 6 are only available if the output file XVSTATS has been specified.
Option 7 is not available if multiple fields are being cross-validated ie if there is more than one record in the estimation parameter file.
Input Files
Name |
Description |
I/O Status |
Required |
Type |
IN |
Input sample data to be cross validated. This must contain X,Y and Z fields and at least one grade field. |
Input |
Yes |
Undefined |
SRCPARM |
Search volume parameter file. This contains 24 compulsory fields defining the search volume and the number of samples needed for grade estimation. More than one search volume may be defined. All fields are numeric:
|
Input |
Yes |
Undefined |
ESTPARM |
Estimation parameter file. Each record in the file describes an estimation method and its associated parameters. The fields are dependent on the estimation methods selected. General fields:
Fields for IPD:
Fields for general case:
Fields for simple kriging:
|
Input |
Yes |
Undefined |
VMODPARM |
Variogram model parameter file. Each record in this file defines a variogram model type and its parameters. Only the VREFNUM field is compulsory.
|
Input |
No |
Variogram - Model |
VGRAM |
Experimental variogram file, as created by the variogram calculation process VGRAM. This experimental variogram file will have been used by the variogram fitting process VARFIT in order to derive the variogram model defined by VMODPARM . This is only required if you want to use access the variogram display and fitting process VARFIT from within XVALID. |
Input |
No |
Variogram - Experimental |
Output Files
Name |
I/O Status |
Required |
Type |
Description |
XVSAMPS |
Output |
No |
Undefined |
Cross-validated output sample file. This contains all the fields from the IN sample data file, plus each grade estimate and associated secondary fields such as kriged variance. |
XVSTATS |
Output |
No |
Undefined |
Output file containing a summary of the input parameters and the cross-validation statistics. It includes a single record for each estimate. The 23 fields in the file are summarised below. If the file already contains all 23 fields then additional records are appended to the file. If the file does not contain all 23 fields, or if the file does not exist, then a new file is created.
|
SAMPOUT |
Output |
No |
Undefined |
Output sample file containing details of weights for each sample for each estimate. |
Fields
Name |
Description |
Source |
Required |
Type |
Default |
X |
X coordinate of sample data in IN file. If not specified, then X is assumed. |
IN |
Yes |
Numeric |
Undefined |
Y |
Y coordinate of sample data in IN file. If not specified, then Y is assumed. |
IN |
Yes |
Numeric |
Undefined |
Z |
Z coordinate of sample data in IN file. If not specified, then Z is assumed. |
IN |
Yes |
Numeric |
Undefined |
ZONE1_F |
First field for zonal control. The field must exist in the IN file and in the ESTPARM file. |
IN, ESTPARM |
No |
Any |
Undefined |
ZONE2_F |
Second field for zonal control. The field must exist in the IN file and in the ESTPARM file. |
IN, ESTPARM |
No |
Any |
Undefined |
KEY |
Key field used to specify the field limiting the number of samples for estimation. The field must exist in the IN file. |
IN |
No |
Numeric |
Undefined |
LENGTH_F |
Field used for length weighting in IPD. The field must exist in the IN file. |
IN |
No |
Numeric |
Undefined |
DENS_F |
Field used for density weighting in IPD. The field must exist in the IN file. |
IN |
No |
Numeric |
Undefined |
Parameters
Name |
Description |
Required |
Default |
Range |
Values |
||||||
SMINFAC |
Multiplying factor which is applied to the first search volume, and used to calculate the exclusion volume for estimation. Samples lying within the exclusion volume are not used for the estimation. The factor must be greater than 0 and less than 1. The exclusion volume is concentric with the search volume. |
No |
0.0001 |
0,1 |
Undefined |
||||||
|
Display control:
|
No |
0 |
0,1 |
0,1 |
Example
!XVALID &IN(holes.d), &SRCPARM(spar10), &ESTPARM(epar10),
&VMODPARM(vpar10), &VGRAM(vgram),&XVSAMPS(xvsamps),
&XVSTATS(xvstats),&SAMPOUT(sampout),@SMINFAC=0.00001,@PRINT=0,@ECHO=0
Related topics and activities
-
ESTIMATE Process