update the normal option for pyxtal

pyxtal/__init__.py

@@ -798,9 +798,7 @@ def subgroup_once(self, eps=0.1, H=None, perms=None, group_type='t', \
  #print("Test subgroup_once", self.group.number, idx, sites)
 
  if idx is None or len(idx) == 0:
- msg = "Cannot find the valid splitter id\n"
- msg += "This is most likely due to the cell size is not sufficient for k-subgroup\n"
- msg += "You may increase the max_cell"
+ msg = "Cannot find valid splitter id (likely due to insufficient cellsize)"
  print(msg)
  return None
 
@@ -3523,7 +3521,8 @@ def get_xtal_string(self, dicts=None, header=None):
  return strs
 
 
- def get_tabular_representation(self, ids=None, normalize=True, N_wp=6, max_abc=50.0, max_angle=np.pi):
+ def get_tabular_representation(self, ids=None, normalize=True,
+ N_wp=6, max_abc=50.0, max_angle=np.pi):
  """
  Convert the xtal to a 1D reprsentation organized as
  (space_group_number, a, b, c, alpha, beta, gamma, wp1, wp2, ....)
@@ -3563,15 +3562,16 @@ def get_tabular_representation(self, ids=None, normalize=True, N_wp=6, max_abc=5
  count = N_cell
  for id, site in zip(ids, self.atom_sites):
  rep[count] = site.wp.index
- if normalize: rep[count] /= len(self.group)
+ if normalize:
+ rep[count] /= len(self.group)
  xyz = site.coords[id] # position
  xyz -= np.floor(xyz)
  rep[count+1:count+4] = xyz
  count += 4
  return rep
 
 
- def from_tabular_representation(self, rep, max_abc=50.0, max_angle=180):
+ def from_tabular_representation(self, rep, max_abc=50.0, max_angle=180, normalize=True):
  """
  Reconstruc xtal from 1d tabular_representation
  Currently assuming the elemental composition like carbon
@@ -3582,20 +3582,31 @@ def from_tabular_representation(self, rep, max_abc=50.0, max_angle=180):
  max_angle (float): maximum angle in radian (used in normalization)
 
  """
- number = int(np.round(rep[0] * 230))
+ if normalize:
+ number = int(np.round(rep[0] * 230))
+ else:
+ number = int(rep[0])
+
  if 1 <= number <= 230:
  group = Group(number)
  [a, b, c, alpha, beta, gamma] = rep[1:7]
- a, b, c = a * max_abc, b * max_abc, c * max_abc
- alpha, beta, gamma = alpha*max_angle, beta*max_angle, gamma*max_angle
+ if normalize:
+ a, b, c = a * max_abc, b * max_abc, c * max_abc
+ alpha, beta, gamma = alpha*max_angle, beta*max_angle, gamma*max_angle
+ else:
+ alpha, beta, gamma = np.degrees(alpha), np.degrees(beta), np.degrees(gamma)
+
  lattice = np.array([a, b, c, alpha, beta, gamma])
  sites_info = np.reshape(rep[7:], (int((len(rep)-7)/4), 4))
  sites = []
  numIons = 0
  for site_info in sites_info:
  [id, x, y, z] = site_info
  if min(site_info) > -0.01:
- wp_id = int(np.round(len(group)*id))
+ if normalize:
+ wp_id = int(np.round(len(group)*id))
+ else:
+ wp_id = int(id)
  if wp_id >= len(group): wp_id = -1
  wp = group[wp_id]
  xyz = wp.search_generator([x, y, z])#; print(wp.get_label(), xyz)
@@ -3607,7 +3618,7 @@ def from_tabular_representation(self, rep, max_abc=50.0, max_angle=180):
  try:
  self.build(group, ['C'], [numIons], lattice, [sites])
  except:
- print("Invalid Build", lattice, numIons, sites)
+ print("Invalid Build", number, lattice, numIons, sites)
  self.valid = False
 
  def get_Pearson_Symbol(self):