summation of mean volume for conductivity calculation for dtype identical

Author: BaerCMI

kinisi/analyzer.py

@@ -156,6 +156,7 @@ def _from_xdatcar(
             ]
             p = u[0]
             p.displacements = sc.concat([i.displacements for i in u], 'particle')
+            p._volume = sc.concat([i._volume for i in u], 'repeats')
             return cls(p)
         elif dtype == 'consecutive':
             structures = _flatten_list([x.structures for x in trajectory])
@@ -226,6 +227,7 @@ def _from_universe(
             ]
             p = u[0]
             p.displacements = sc.concat([i.displacements for i in u], 'particle')
+            p._volume = sc.concat([i._volume for i in u], 'repeats')
             return cls(p)
         else:
             raise ValueError('The dtype specified was not recognised, please consult the kinisi documentation.')
@@ -281,6 +283,7 @@ def _from_ase(
             ]
             p = u[0]
             p.displacements = sc.concat([i.displacements for i in u], 'particle')
+            p._volume = sc.concat([i._volume for i in u], 'repeats')
             return cls(p)
         elif dtype == 'consecutive':
             structures = _flatten_list([x for x in trajectory])

kinisi/diffusion.py

@@ -219,6 +219,8 @@ def _conductivity(
         _conductivity_diffusion_coefficient = sc.to_unit(
             self.gradient / (2 * self.dg['dimensionality'].value), 'coulomb2cm2/s'
         )
+        if volume.shape[0] >= 1:
+            volume = sc.sum(volume)     
         conversion = number_of_particles / (volume * k * temperature)
         _sigma = sc.to_unit(_conductivity_diffusion_coefficient * conversion, 'mS/cm')
         self._sigma = Samples(_sigma.values, _sigma.unit)

kinisi/parser.py

@@ -116,7 +116,7 @@ def __init__(
         self.dimensionality = drift_corrected.sizes['dimension'] * sc.units.dimensionless
 
         self.displacements = drift_corrected['particle', indices]
-        self._volume = np.prod(latt.values[0].diagonal()) * latt.unit**3
+        self._volume = np.mean([np.prod(latt.values[i].diagonal()) for i in range(latt.sizes['time'])]) *  latt.unit**3
 
     def _to_datagroup(self, hdf5=True) -> sc.DataGroup:
         """