Fcidump from MRCC and PySCF

Hi,

  I did a comparison between the fcidump files from MRCC and PySCF. It seems there are sign differences between two-electron integrals and differences in absolute values in one-electron integrals.
Namely, pyscf fcidump

 -0.3164468354453429    1    1    2    1
 -1.941012551303533    1    1  0  0

fort.55

  0.31644683544545071374E+00     2     1     1     1
 -0.91414792593896876394E+00     1     1     0     0

 

The sign difference may come from phases in orbitals. But the absolute value difference is harder to trace. I think pyscf one-electron integral is T + Ven. May I know the possible reason for this differences? The converged energies seem similar.Thanks a lot.

Attached input, output, and fcidump files.

By default, Fock-matrix elements are written to file fort.55. Please set dens=2, then core Hamiltonian elements will be printed.

Thank you so much! I added dens=2 in the input, and found some extra lines in fort.55namely, (I suppose the lines above  1 2 3 4 5 are integrals with core Hamiltonian)

    1     2     3     4     5     1     2     3     4     5 
0.00000000000000000000E 00     1     1     0     0 
0.00000000000000000000E 00     1     2     0    
0  0.00000000000000000000E 00     1     3     0     0  0.00000000000000000000E 00     1     4     0     0 
0.48331367864476776353E 00     1     5     0     0 
0.00000000000000000000E 00     2     1     0    
0  0.00000000000000000000E 00     2     2     0     0  0.00000000000000000000E 00     2     3     0     0 
0.00000000000000000000E 00     2     4     0     0 
-0.14900273403827146712E 00     2     5     0     0 
0.00000000000000000000E 00     3     1     0     0
...

what does these lines mean? updated input and fort.55 files are attached.

 

These are dipole moment integrals. You can find the one-electron integrals above those lines.

Thank you so much!!