zfbi/rgtNet: Deep learning for simultaneously interpreting 3D seismic horizons and faults by estimating a relative geologic time volume

RgtNet:using synthetic datasets to train an end-to-end CNN for 3-D RGT(Relative Geologic Time) estimation

This is a Pytorch version of RgtNet for 3-D RGT(Relative Geologic Time) estimation

Getting Started with Example Model for RGT estimation

If you would just like to try out a pretrained example model, then you can download the pretrained model [neuc] and use the demo.ipynb script to run a demo (example data can be downloaded from here).

Requirments

python>=3.6
torch>=1.0.0
torchvision
torchsummary
natsort
numpy
pillow
plotly
pyparsing
scipy
scikit-image
sklearn
tqdm

Install all dependent libraries:

pip install -r requirements.txt

Dataset

To train our CNN network, we automatically created 400 pairs of synthetic seismic and corresponding RGT volumes, which were shown to be sufficient to train a good RGT estimation network.

The training and validation datasets can be downloaded here

Training

Run train.sh to start training a new RgtNet model by using the synthetic dataset

sh train.sh

Validation & Application

Run infer.sh to start applying a new RgtNet model to the synthetic or field seismic data

sh infer.sh

License

This extension to the Pytorch library is released under a creative commons license which allows for personal and research use only. For a commercial license please contact the authors. You can view a license summary here: http://creativecommons.org/licenses/by-nc/4.0/