Deep Galerkin Method

Deep Galerkin method is a deep learning framework for solving partial differential equations. This particular implementation can be run, straight out-of-the-box in Docker containers. It is also highly customizable.

Install

gh repo clone ShuAiii/DGM

Build and run Docker container

make build_dgm will pull the base tensorflow image and build the DGM source over the base image. make dgm will run a container of the image with gpu capability, make dgm_cpu will run a cpu only container.

Example

Black-Scholes call option

Inside the container, run:
python train.py