1D Lattice Structure with indices

[flipdazed]

Issue A tensor format was adopted early on as initial structures were 2D and the thought was that operations might be non-local e.g. Einstein summation across nd-tensors

This is not the case in fact all operations are local and only depend on the neighbours and periodic boundaries so a 1D array suffices and operations need not be more complex across non local neighbours save the trivial boundary conditions

Resolution

• remove the Lattice() structure
• replace with functions on neighbours in a 1D array with periodic boundary

Places with Periodic_Lattice references

• hmc.dynamics.Leap_Frog._moveX() / _moveP()
• hmc.potentials.Shared._*// Klein_Gordon / Quantum_Harmonic_Oscillator
• hmc.lattice.Periodic_Lattice
• hmc.hmc.Hybrid_Monte_Carlo.__init__
• test.*
• results.*

[flipdazed]

done lattice.py and tested with test_lattice

[flipdazed]

done dynamics.py and tested with test_dynamics

[flipdazed]

done test_potentials

[flipdazed]

done hmc.Hybrid_Monte_Carlo.__init__()

[flipdazed]

done all but results

[flipdazed]

done results