SWT_ML

Surface Wave Tomography using Machine Learning

Contents

rawData: original data downloaded from IRIS

• data: scripts to convert information in rawData to ML input (inputs are also saved here temporarily)

  • syn_disp_from_vs: calculating synthetic dispersion based on vs profile
  • dataPrepare.py: preparing training dataset(US) and testing dataset(CN)

• train: scripts to do training and predicting

  • Main_Train.py: run training (set self.pretrained = False, self.start = 0 in config.py before training)
  • Main_Predict.py: predicting (set self.pretrained = True, self.start = 600 in config.py before predicting)
  • checkLayer.py: plot a map to check result

Work flow

1. data/dataPrepare.py, four parts separated by docstring
2. Main_Train.py
3. Main_Predict.py