QuantumLab-ZY/HamGNN

Issues

• Incompatible basis set for universal Hamiltonians model

  #31 opened 16 days ago by ChenQianA
  1

• How to get the total energy of my structure?

  #29 opened a month ago by ayitido
  1

• requirements for conda environment?

  #25 opened a month ago by miicck
  6

• Some problems with using HamGNN for the first time

  #27 opened a month ago by ayitido
  6

• Prediction of overlap matrix.

  #28 opened a month ago by flamingoXu
  3

• openMX_postprocess installation error static declaration of ‘Print_CubeCData_MO’

  #26 opened 2 months ago by Vahid999
  4

• setup.py error

  #24 opened 3 months ago by Drawthe
  2

• Several question about training

  #23 opened 3 months ago by liangzhixin-202169
  2

• Excessive Memory Usage in `band_cal_parallel.py` for Large Systems

  #22 opened 3 months ago by newplay
  4

• Python environment for HamGNN

  #14 opened 6 months ago by flamingoXu
  9

• Preparing multi materials dataset

  #21 opened 5 months ago by Youhaojen
  10

• RuntimeError: CUDA out of memory.

  #20 opened 5 months ago by Youhaojen
  2

• The MKL dependency error in band_cal_parallel.py

  #12 opened 6 months ago by newplay
  4

• Will HamGNN support fitting the TB Hamiltonian generated by ABACUS in the future?

  #2 opened a year ago by sgyang345
  5

• the format of prediction_hamiltonian.npy

  #19 opened 5 months ago by flamingoXu
  1

• RuntimeError with band_cal_parallel

  #18 opened 6 months ago by flamingoXu
  5

• The usage of graph_data_gen

  #17 opened 6 months ago by flamingoXu
  12

• problem：

  #16 opened 6 months ago by flamingoXu
  2

• how to turn on the prediction of wavefunction and overlap

  #15 opened 6 months ago by flamingoXu
  1

• the setting exampleK_path

  #13 opened 6 months ago by flamingoXu
  3

• Problem in the training process when training a model for silicon

  #10 opened 7 months ago by SharpLonde
  11

• Regarding convergence issues during training

  #11 opened 7 months ago by newplay
  14

• Some problem of 2nd training processed

  #9 opened 8 months ago by newplay
  6

• The 'Cuda error: device-side assert triggered' occurred during the training process(exacte error: RuntimeError: Class values must be smaller than num_classes.)

  #8 opened 8 months ago by newplay
  14

• Attempted to utilize the model from the Twist Bilayer MoS2 (TBM) Demo to predict angles for other configurations, but was unsuccessful.

  #7 opened 9 months ago by newplay
  14

• After replacing the demonstration structure graph file with my own structure graph file, 'band_cal' fails to function properly.

  #6 opened 9 months ago by newplay
  3

• Regarding the issue of mismatched shapes encountered during the processing of loading the Hamiltonian matrix.

  #5 opened 10 months ago by newplay
  2

• The limitation in the training process

  #4 opened 10 months ago by newplay
  2

• openmx.c compilation error resolved

  #3 opened a year ago by newplay
  2

• model can not save to ckpt

  #1 opened a year ago by Youhaojen
  2