PEN-FWI network executable files and datasets
- Submission Dates:
-
to
- Citation Author(s):
-
Jiahao Ren
- Submitted by:
- Yang Liu
- Last updated:
- DOI:
- 10.21227/sfmr-bb58
- Data Format:
- Categories:
- Keywords:
Abstract
The dataset introduces a novel physics-embedded deep learning neural network for accelerating traditional FWI algorithms, thereby reducing the required imaging time while overcoming the challenge of needing a high-quality initial model for traditional FWI inversion. The provided dataset includes training, validation, and testing sets, along with executable files related to PEN-FWI network training and validation.
Instructions:
These datasets and network structures have been implemented in a Python environment, with the datasets and Python code integrated into executable files. Users can obtain inversion results by following the steps outlined in the provided "read_me" file.
