promptdojo_promptdojo_promptdojo_promptdojo_promptdojo_promptdojo_promptdojo_promptdojo_promptdojo_promptdojo_promptdojo_

Pipeline orchestration shape without the theater — step 7 of 7

ch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelinesch 37 · dataset formats, ingestion, and validation pipelines7/7
Checkpoint

One last thing before we move on. pass this to mark the lesson done, or skip and keep moving. hop to the next when you're ready.

Prove restartability: run clean once and persist its output in OUTPUTS["clean"]. Run feature_v1 (which has a bug — it forgets to cast amt to int), then 'fix the bug' by running feature_v2 FROM THE PERSISTED clean output — without re-running clean. Print the final features, then how many times each stage ran (clean: 1, feature: 2).

full-screen editor opens — close anytime to keep reading.

open the editor to run it.