Batch mode runs many target/proteome combinations from a manifest.
thyme-batch generate-manifest \
--base-config pipelines/thyme/batch_setup/batch_example/example_base_config.yaml \
--targets pipelines/thyme/batch_setup/batch_example/targets_tiny.csv \
--proteomes pipelines/thyme/batch_setup/batch_example/proteomes_tiny.csv \
--batch-root /tmp/thyme_batchtargets.csv requires target_id. proteomes.csv requires proteome_id. Additional columns are dot-path config overrides, for example:
target_protein.structuredb_idproteome.sourceproteome.proteome_fasta_pathdecoder.beam_width
thyme-batch run \
--manifest /tmp/thyme_batch/manifest.csv \
--max-workers 4 \
--resume
thyme-batch summarize \
--manifest /tmp/thyme_batch/manifest.csvEach row gets its own config, output directory, stdout/stderr logs, status JSON, and metrics artifacts.
When --wandb-project is provided, batch mode creates one W&B run per manifest row. Logged fields include target/proteome/config IDs, pipeline git SHA, top scores, diversity, runtime, decode success, sequence counts, and artifacts.