mllf API

File Handling

For detailed documentation on file formats and usage examples, see File Handling.

`mllf.file_handling.read_bias_coeff`
`mllf.file_handling.read_rtf`	RTF file parser
`mllf.file_handling.read_pdb`	PDB file parser
`mllf.file_handling.write_bias_coeff`	Write bias coefficient files in the old .inp or .py style used by the simulator.
`mllf.file_handling.read_output`
`mllf.file_handling.generate_combinations`	Generate all combinations of site/sub files into separate directories.

For detailed documentation on DeepSet pretraining, see AtomBondGNN Pretraining.

`mllf.cb.deepset_autoencoder`	DeepSet Autoencoder for pretraining atom-level embeddings.
`mllf.cb.deepset_pretraining_dataset`	Dataset generation for DeepSet autoencoder pretraining.
`mllf.cb.train_deepset_autoencoder`	Training script for DeepSet autoencoder pretraining.
`mllf.cb.aev_processor`

For detailed documentation on CB architecture, see Contextual Bandit Setup.

`mllf.cb.graph`	Graph structure for contextual bandit.
`mllf.cb.policy`	Edge value policy built on top of a node encoder.
`mllf.cb.value_net`	Value network for baseline estimation in REINFORCE.
`mllf.cb.rgcn`	RGCN encoder using PyTorch Geometric.
`mllf.cb.graph_utils`	Helpers to convert existing Graph objects to PyTorch Geometric Data.

For detailed documentation on the workflow system, see Workflow System.

High-level workflow utilities for preparing combos, training and running sims.