domains / families / science / bionemo-kermt-monitor

bionemo-kermt-monitor

separate

Check progress for a detached KERMT run (pretrain, finetune, or any kermt_run_detached invocation). Reads run.json, queries docker for container state, tails the pretrain/finetune log, and parses progress lines (epoch, step, val loss).

When to use

• Check progress for detached KERMT runs
• Query docker for container state
• Tail pretrain or finetune logs
• Parse progress lines like epoch, step, and val loss
• Stream live logs with --follow
• Emit structured status report with --json
• Display final test metrics for completed finetune runs

When not to use

• Blocking KERMT runs like kermt-infer or kermt-embed
• Modifying run.json or checkpoint directories
• Killing containers without explicit user instruction
• Pulling or modifying the kermt image

Skill page

Source