dhudsmith/qmnnet

Teaching a machine to rapidly solve the 1-d single-particle Schrodinger equation.

qmnnet

qmnnet uses a neural network to rapidly solve the 1D Schrodinger equation.

Structure

qmnnet involves three basic logical steps:

1. Generating an ensemble of random poentials
2. Solving Schrödinger's equation for the eigenvalues and eigenvectors for each potential
3. Train a neural network to predict the eigenvalues for new potentials

The jupyter notebook `qmnnet.ipynb' goes through these steps. Click the Binder badge to run the jupyter notebook in your browser.