CLI¶

AdaptivePy provides a Click-based command-line interface installed as adaptivepy.

Commands¶

Execute a full adaptive sampling workflow from a YAML configuration:

adaptivepy run config.yaml

This loads features, clusters frames, applies all configured policies, selects seeds, and writes outputs to output_dir.

Check configuration and input data without running clustering:

adaptivepy validate config.yaml

Useful for catching missing files, shape mismatches, or invalid policy names before a long run.

Print all registered policy names:

adaptivepy list-policies

Example output:

fast
knn_as
least_counts
ma_reap
maxent_vampnet
random
ts_dar

Built-in policies:

Policy	Description
`least_counts`	Select least-populated clusters
`random`	Random cluster sampling
`fast`	Feature-directed exploration/exploitation (Zimmerman & Bowman 2015)
`ma_reap`	Multiagent REAP with agent stakes and learned CV weights (Kleiman & Shukla 2022)
`knn_as`	k-nearest-neighbors adaptive sampling over cluster representatives (Rovers et al. 2025)
`maxent_vampnet`	Entropy-based frame selection via VAMPNet; no clustering (Kleiman & Shukla 2023)
`ts_dar`	OOD-score frame selection via TS-DAR hyperspherical embeddings (Liu et al. 2025)

See Policies for configuration details.

adaptivepy --version

Code	Meaning
0	Success
1	Validation or runtime error

Errors are printed to stderr with a short message.

Feature-only run:

adaptivepy run examples/config.yaml

Validate before running:

adaptivepy validate examples/config.yaml && adaptivepy run examples/config.yaml