TorchTDSCHA documentation

TorchTDSCHA implements the time-dependent extension of the stochastic self-consistent harmonic approximation (SSCHA), enabling the real-time propagation of quantum nuclei in anharmonic potentials. It solves the TDSCHA equations through exact Gaussian integration of the quantum forces, accelerated with PyTorch for massively parallel tensor algebra on GPUs.

API Reference

• init module
• averages module
  • V_classic()
  • av_V()
  • av_V_t()
  • av_d3()
  • d2V()
  • ext_for()
  • f_classic()
  • force()
  • force_t()
  • kappa()
  • kappa_t()
  • torch_av_V()
  • torch_ext_for()
  • torch_force()
  • torch_kappa()
• dynamics module
• phonons module
  • displace_along_mode()
  • get_AB()
  • get_alpha()
  • get_phonons()
  • get_phonons_THz()
  • get_phonons_r()
  • inv_Phi()
  • print_phonons()
  • print_phonons_mat()
  • regularize()
  • remove_translations()
  • remove_translations_from_mat()
• diff_2nd module
• diff_3rd module
• diff_4th module