On Apr 8, 2016, at 9:01 AM, Forwordom <mljphy@qq.com> wrote:

Dear, M. Dolfi

I am a freshman about DMRG. I have two questions :
(1) How to get truncation error using mps_optim codes ?

An example is given in the tutorial files that you find in /tutorials/mps-01-optim.

iterations = pyalps.loadIterationMeasurements(pyalps.getResultFiles(prefix='parm_spin_one_half’), what=['TruncatedWeight'])
truncation_iteration = pyalps.collectXY(pyalps.flatten(iterations), 'iteration', 'TruncatedWeight')

(2) How to get von Neumann entanglement entropy S(A) of a system A.
Actually, I have obtained entanglement entropy data by setting MEASURE[Entropy]=1 using mps_optim. For a spin chain, we got the data number of entanglement entropy which seems the same as the number of 'bond' in chain. For example, to run a mps_optim of a spin chain with 32 sites , I will get 31 data in output of entropy. What meaning of such entanglement entropy data?
How to get von Neumann entanglement entropy S(A) of a two dimension system A (Just like page 5 in paper Phys.Rev.B 64,024424(2012))

The Entropy measurement is the von Neumann entanglement entropy, computed on each bond.

In the paper that you highlight, I think the authors optimized one MPS for each L_y, then compute the von Neumann entanglement entropy on the central MPS bond.

Basically on a L_x * L_y = 20 * 4 system, on the bond between the MPS site 39 and 40. (when counting from 0)

Best regards,

Michele