make input consistent with CPC paper data

input/cobalt-adatom/generate_cobalt_adatom_input.py

@@ -0,0 +1,72 @@
+#!/usr/bin/env python
+
+import h5py
+import numpy as np
+
+# We use pytriqs package for coulomb matrix calculation
+import pytriqs.operators.util as U_matrix
+
+
+# Coulomb matrix
+U = np.real(U_matrix.U_matrix(2, U_int=6.6, J_hund=0.9, basis='cubic'))
+# Chemical potential
+xmu = 44.44
+
+# Impurity energies
+Eps0 = np.array([[-0.70,-0.70], [-0.8,-0.8], [-0.50,-0.50], [-0.8,-0.8], [-0.70,-0.70] ])
+# Hybridization between impurity and bath
+Vk =   [ np.array([ [0.56434,0.56434],[0.68392,0.68392],[0.29519,0.29519]]),
+         np.array([ [0.81892,0.81892],[0.99136,0.99136]]),
+         np.array([ [0.77347,0.77347],[0.79785,0.79785]]),
+         np.array([ [0.81892,0.81892],[0.99136,0.99136]]),
+         np.array([ [0.56434,0.56434],[0.68392,0.68392],[0.29519,0.29519]]) ]
+# Bath energies
+Epsk = [ np.array([ [-2.37325, -2.37325],[-0.87328,-0.87328], [2.01265,2.01265]]),
+         np.array([ [-3.15496,-3.15496],[-1.69066,-1.69066] ]),
+         np.array([ [-5.59842,-5.59842],[-2.95325,-2.95325]]),
+         np.array([ [-3.15496,-3.15496],[-1.69066,-1.69066]]),
+         np.array([ [-2.37325, -2.37325],[-0.87328,-0.87328], [2.01265,2.01265]] ) ]
+
+# Total number of fermionic orbitals
+Ns = 17
+# Number of impurity orbitals
+ml = 5
+
+# Interorbital hoppings
+t0 = [ np.array([[0, 0]]*ml),
+       np.array([[0, 0]]*ml),
+       np.array([[0, 0]]*ml),
+       np.array([[0, 0]]*ml),
+       np.array([[0, 0]]*ml) ]
+
+# (n_up, n_down) symmerty sectors
+sectors = np.array([[12,15],[15,12],[14,13],[13,14],[13,15],[14,14],[15,13],])
+
+# open data file
+data = h5py.File("input.h5", "w");
+# Inverse temperature
+#beta = data.create_dataset("BETA", shape=(), dtype='f', data=10.0)
+
+hop_g = data.create_group("sectors")
+hop_g.create_dataset("values", data=sectors)
+
+bath = data.create_group("Bath")
+
+for i in range(ml):
+  if(Epsk[i].shape != Vk[i].shape):
+    raise "Incorrect shape for Hybridisation and Epsk"
+  Epsk_g = bath.create_group("Epsk_" + str(i))
+  Epsk_g.create_dataset("values", shape=(len(Epsk[i]),2,), data=Epsk[i], dtype=np.float)
+  Vk_g = bath.create_group("Vk_" + str(i))
+  Vk_g.create_dataset("values", data=np.array(Vk[i]), dtype=np.float)
+  t0_g = data.create_group("t0_" + str(i))
+  t0_g.create_dataset("values", data=np.array(t0[i]), dtype=np.float)
+
+hop_g = data.create_group("Eps0")
+hop_g.create_dataset("values", data=Eps0)
+
+int_g = data.create_group("interaction")
+int_ds = int_g.create_dataset("values", shape=(ml,ml,ml,ml,), data=U)
+
+int_ds = data.create_dataset("mu", shape=(), data=xmu)
+

input/hubbard-16site/generate_16site_plaquette_input.py

@@ -0,0 +1,58 @@
+#!/usr/bin/env python
+
+import h5py
+import numpy as np
+
+U = np.array([6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0, 6.0])
+xmu = np.array([0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54, 0.54])
+
+t = -1.0;
+tp = 0.3;
+
+sectors = np.array([[6,6]])
+
+
+Nrows = 4;
+Ncols = 4;
+
+tm = np.zeros((Nrows*Ncols, Nrows*Ncols));
+
+for row in range(0, Nrows):
+  for col in range(0, Ncols):
+    i = row * Ncols + col;
+    # Horizontal hopping with period.
+    j = row * Ncols + (col + 1)%Ncols;
+    tm[i, j] = -t;
+    tm[j, i] = -t;
+    # Vertical hopping with period.
+    j = ((row + 1)%Nrows) * Ncols + col;
+    tm[i, j] = -t;
+    tm[j, i] = -t;
+    # Diagonal (\) hopping with period.
+    j = ((row + 1)%Nrows) * Ncols + (col + 1)%Ncols;
+    tm[i, j] = -tp;
+    tm[j, i] = -tp;
+    # Diagonal (/) hopping with period.
+    j = ((row + 1)%Nrows) * Ncols + (col - 1)%Ncols;
+    tm[i, j] = -tp;
+    tm[j, i] = -tp;
+
+print "bandwidth: %f" % (np.amax(np.linalg.eig(tm)[0]) - np.amin(np.linalg.eig(tm)[0]));
+
+
+Ns = len(U)
+
+data = h5py.File("input.h5", "w");
+
+hop_g = data.create_group("sectors")
+hop_g.create_dataset("values", data=sectors)
+
+
+hop_g = data.create_group("hopping")
+hop_g.create_dataset("values", data=tm)
+
+int_g = data.create_group("interaction")
+int_ds = int_g.create_dataset("values", shape=(Ns,), data=U)
+
+int_g = data.create_group("chemical_potential")
+int_ds = int_g.create_dataset("values", shape=(Ns,), data=xmu)