Hi all, I have coded a ferret script where 2 variables are co-plotted with coloring samples with their latitudes. It can be easilly achieved with the script polymark. yes? go polymark polygon/over/nolab/key/lev=(-90,90,10)/pal=centered/nolab/fill w1,w2,w3,circle,0.5 No problem there. See 1st figure attached. But as I worked from 2D variables, when I vectorize them, I conserve their underlaying grid. For now, I have done this using an external call to python/numpy yes? go shuffle 10 yes? list shuffle VARIABLE : SHUFFLE FILENAME : shuffle.txt SUBSET : 10 points (X) 1 / 1: 3.00 2 / 2: 10.00 3 / 3: 8.00 4 / 4: 4.00 5 / 5: 7.00 6 / 6: 6.00 7 / 7: 5.00 8 / 8: 1.00 9 / 9: 2.00 10 / 10: 9.00 Applying this shuffle to my samples with the ferret samplei(v1, shuffle) gives me what I was expecting (independant sample over their latitudes). yes? go polymark polygon/over/nolab/key/lev=(-90,90,10)/pal=centered/nolab/fill samplei(w1,shuffle),samplei(w2,shuffle),samplei(w3,shuffle),circle,0.5 See fig2 attached. I am wondering if a shuffle function could be added to ferret ? Any news to share about link between python and ferret ? Patrick
Attachment:
ScatterPlot_02.gif
Description: GIF image
Attachment:
ScatterPlot_03.gif
Description: GIF image
#!/bin/env python import numpy, sys a=numpy.random.permutation(int(sys.argv[1])) for i in range(len(a)): print a[i]+1
Attachment:
shuffle.jnl
Description: application/jnl