NOTICE: This version of the NSF Unidata web site (archive.unidata.ucar.edu) is no longer being updated.
Current content can be found at unidata.ucar.edu.
To learn about what's going on, see About the Archive Site.
Dear all: i have similar problem to yours with the full description below: below are sample lines in my file: private static final int NUM_SPIKES_X = 4; // number of spikes along Y axis private static final int NUM_SPIKES_Y = 3; // increase density value to make spikes "thinner" // decrease density value to make spikes "thicker" private static final int DENSITY = 8; and then i construct the integer2Dset with the density and spikes_Y and X using the constructor Integer2DSet set = new Integer2DSet( DENSITY * NUM_SPIKES_X, DENSITY * NUM_SPIKES_Y); then i have the two for-loops to read the spike value and put on the grid.i think that is causing the scale mis-match,each oject carrying the spike value also carries a pi value and a mass value.but this values are not mapped to the x and Y axis as it is supposed to. String name=(String)pro.getName(); double ratiox=pro.getProp(); double massx=pro.getMass(); double pix = pro.getPi(); so the domain gat to be numbers like massx and pix values taken from the prodat oject instead of the DENSITY * NUM_SPIKES_X, DENSITY * NUM_SPIKES_Y so it is clear that, if there is a way to pass the mass value and the pi value through the integer2Dset constructor so that each spike will correspond to a mass value and a pi value in the domain.as it stands now, the domain is independent and NOT connected to the range logically.The data is not fully represented very well.so i am looking for ways to include the mass and pi values in the domain on the grid so that the scale for the pi and mass will match accurately.you can test it using the sample below: sample data below: Name mass pi ratio ATF4_HUMAN 38534 4.83 1 ATHL_HUMAN 115437 6.12 .7352244077830988 ATNB_HUMAN 35039 8.74 .5275324058321538 ATND_HUMAN 31492 8.58 .6540817256804714 ATPW_HUMAN 15691 5.85 .44756554307116103 ATS1_HUMAN 103431 6.27 .415245531128005 ATX7_HUMAN 95392 9.89 .36708860759493667 AXO1_HUMAN 113323 8.11 .5319148936170213 BAC1_HUMAN 801906 4.95 .6578947368421053 BAD_HUMAN 18397 10.1 .5154639175257733 BASI_HUMAN 29203 5.41 .5952380952380952 BASI_HUMAN 29203 5.41 .542363203491572 BASO_HUMAN 110909 6.91 .6711409395973155 BB1_HUMAN 6587 7.85 1 BC10_HUMAN 902869 6.21 .10714285714285712 Everything about the display is fine except this issue of domain and scale. it will be good to know some workarounds on this or any constructor that can implement this. thanks for all your time Isaac -----Original Message----- From: Bill Hibbard To: z karas Sent: 9/16/02 12:52 PM Subject: Re: build doman_set ? z karas wrote: > > Hi > > My coordinates X, Y are stored in a 2D array: coord. > > coord[2][1000]; > coord[0][0]= x1,coord[0][1] > x2,...,coord[0][999]=x1000 > coord[1][0]= y1,coord[1][1]=y1,...,coord[1][999]=y1000 > > how I can build a domain_set from this array ? If your points are arranged in a curved 2-D grid, then use the Gridded2DSet constructor: public Gridded2DSet(MathType type, float[][] coord, int lengthX, int lengthY) whree lengthX * lengthY = 1000. Otherwise use the Irregular2DSet constructor: public Irregular2DSet(MathType type, float[][] coord) Good luck, Bill
visad
archives: