The P2B1 capability (P2B1) is an autoencoder that determines a set of features to most efficiently describe molecular dynamics (MD) simulation data.
Scientists interested in working with efficient representations of MD simulation data.
Scientists can train the model on their own data and use the resulting reduced set of features as input for further analysis.
MD simulation data consist of many descriptors. This capability shows how you can use an autoencoder to compress these descriptors into a minimal set that faithfully describes the data. This enables downstream analysis using a more tractable dataset as input.
The following components are in the Model and Data Clearinghouse (MoDaC):
- Data:
- The default dataset is 3k disordered 3-component-system (DPPC-DOPC-CHOL).
- Converged Model:
- The trained weights (for both the full model and just the encoder;
.hdf5files) and the corresponding model topologies (.jsonfiles) are stored in Autoencoder for MD Simulation Data (P2B1).
- The trained weights (for both the full model and just the encoder;
Refer to this README.