Variational Monte Carlo for 1D Heisenberg model

using Haldane-Shastry variational wavefunction

Compile as: make vmc

To run: ./vmc.x [Char(20) UUID of the run] [Int seed]

Example: ./vmc.x 'testRun' 3127387 < model.in > sim.out

Starts simulation for 1D Heisenberg system of L sites as given in model.in. The simulation performs nOPT iterations of optimization cycle, during which the new updated variational parameters p are computed. The magnitude of update is proportional to value of opt_eps. Each single iteration consists of nEQ equilibration sweeps and nPROD production sweeps for averaging. The data from optimization are stored in testRun-opt.dat file.

If the value of nOPT= 1, the samples of energy and spin-spin correlation functions are saved to files testRun-vals.dat and testRun-corr.dat respectively for subsequent analysis. The average values of energy and spin-spin correlation functions are given in sim.out as well.

Data analysis:

Sample means

bining.py computes average and estimates the error of sampled observable by decorrelating the data using binning analysis

To run: python3 binning.py [String filename] [Int column] [Int #samples] [Int MaxBinSize]

Example: python3 binning.py 'testRun-vals.dat' 1 1000 50 > averages.dat

Performs binning analysis of samples of energy stored in column No. 1 (starting from 0th column)

References

[1] A. W. Sandvik, arXiv:1101.3281

[2] S. Sorella, arXiv:cond-mat/0502553