A tuned inspection model, refined on improved training data.
What it does
Delivers a refined inspection-model checkpoint by chaining baseline AutoML tuning, DEFT data improvement, and a final AutoML sweep.
When to use it
Use for an end-to-end AOI training workflow or requests to chain AutoML and DEFT; exclude requests for only one stage.
How to use it
Give it an inspection-model training workspace; it runs the three approved training phases and produces a winning checkpoint.
What you provide
- An existing project
Good to know
- acts without asking
What you need · 4
Requires Docker; the delegated workflows declare additional requirements.
Requires NVIDIA Container Toolkit alongside Docker.
Runs the AutoML skill for baseline tuning and final refinement; these delegated phases are required to complete the pipeline.
Requires this DEFT skill for the default AOI workflow's second phase; non-AOI workflows instead invoke a matching DEFT skill.
Access · 0
No additional actions listed.