Change verbosity

src/libcov.c

@@ -30,15 +30,15 @@ void build_dSdC( int nside, int nstokes, int npix, int nbin, long *ispec, long *
 //   const int nspec = nstokes2nspec(nstokes);
   const int64_t npixtot = npix*nstokes;
 
-  fprintf( stdout, "lmax=%d\n", lmax);
-  fprintf( stdout, "npix=%d\n", npix);
-  fprintf( stdout, "nstokes=%d\n", nstokes);
-  fprintf( stdout, "nspec=%d\n", nspec);
+//   fprintf( stdout, "lmax=%d\n", lmax);
+//   fprintf( stdout, "npix=%d\n", npix);
+//   fprintf( stdout, "nstokes=%d\n", nstokes);
+//   fprintf( stdout, "nspec=%d\n", nspec);
 
   int64_t ntot = nspec*nbin*npixtot*npixtot;
-  fprintf( stdout, "memset (%d,%d,%d,%d = %lld)...\n", nspec,nbin,nstokes,npix,ntot);
+//   fprintf( stdout, "memset (%d,%d,%d,%d = %lld)...\n", nspec,nbin,nstokes,npix,ntot);
   memset( dSdC, 0., ntot * sizeof(double));
-  fprintf( stdout, "ispec = (%d,%d,%d,%d,%d,%d)\n", ispec[0],ispec[1],ispec[2],ispec[3],ispec[4],ispec[5]);
+//   fprintf( stdout, "ispec = (%d,%d,%d,%d,%d,%d)\n", ispec[0],ispec[1],ispec[2],ispec[3],ispec[4],ispec[5]);
 
 #pragma omp parallel default(none) shared(stdout,nbin,nside,npix,nstokes,dSdC,ipix,bl,ellbins,ispec)
   {

xqml/xqml.py

@@ -91,6 +91,14 @@ class xQML(object):
         clth = np.asarray(clth)
         if len(clth) == 4:
             clth = np.concatenate((clth,clth[0:2]*0.))
+
+        nbin = len(self.ellbins)
+        nmem = self.nspec*nbin*(self.nstokes*self.npix)**2
+        toGb = 1024. * 1024. * 1024.
+        print( "xQML")
+        print( "spec: ", spec)
+        print( "nbin: ", nbin)
+        print( "Memset: %.2f Gb (%d,%d,%d,%d)" % (8.*nmem/toGb,self.nspec,nbin,self.nstokes,self.npix))
         
         # If Pl is given by the user, just load it, and then compute the signal
         # covariance using the fiducial model.