Create dataset card

OpenFWI is a collection of large-scale, multi-structural benchmark datasets for machine learning driven seismic FWI. They released twelve datasets synthesized from different priors, including one 3D dataset. They also provided baseline experimental results with four deep learning methods: InversionNet, VelocityGAN, UPFWI and InversionNet3D.

OpenFWI is the first open-source platform to facilitate data-driven FWI research. It will be actively developed and the datasets are expected to evolve.

I parsed all the openFWI dataset files found [here](https://smileunc.github.io/projects/openfwi/datasets) and put them into a huggingface dataset.

If you use OpenFWI Datasets in your work, please cite the below paper (Bibtex below).
```
@inproceedings{NEURIPS2022_27d3ef26,
author = {Deng, Chengyuan and Feng, Shihang and Wang, Hanchen and Zhang, Xitong and Jin, Peng and Feng, Yinan and Zeng, Qili and Chen, Yinpeng and Lin, Youzuo},
booktitle = {Advances in Neural Information Processing Systems},
editor = {S. Koyejo and S. Mohamed and A. Agarwal and D. Belgrave and K. Cho and A. Oh},
pages = {6007--6020},
publisher = {Curran Associates, Inc.},
title = {OpenFWI: Large-scale Multi-structural Benchmark Datasets for Full Waveform Inversion},
url = {https://proceedings.neurips.cc/paper_files/paper/2022/file/27d3ef263c7cb8d542c4f9815a49b69b-Paper-Datasets_and_Benchmarks.pdf},
volume = {35},
year = {2022}
}
```

README.md

@@ -0,0 +1,13 @@
+---
+license: cc-by-4.0
+task_categories:
+- image-segmentation
+- image-to-image
+language:
+- en
+tags:
+- geology
+pretty_name: Geophysical Waveform Inversion Dataset
+size_categories:
+- 100K<n<1M
+---