Milestone map
Milestone map
2 milestones
Do the phase's lessons in public
1–2 weeks (about 10 hrs/week, est)
Work through the 28 lessons of Phase 4: Computer Vision from AI Engineering from Scratch (free) and keep the computer vision lessons' code and notebooks (convolutions, CNNs, transfer learning, detection, segmentation, vision transformers) in one public repo as you go. About 16 hours (est).
Proof required
Link your public repo with the computer vision lessons' code and notebooks (convolutions, CNNs, transfer learning, detection, segmentation, vision transformers), and show your self-check (quizzes passed, notebooks run, tests passing where they exist).
What gets checked
- Each lesson's code runs and its quiz is passed.
- Models are evaluated on held-out images.
- Results include the metric the lesson uses (accuracy, mAP, IoU).
Common mistakes
- Copying the repo's solutions: verifiers read your history and ask about your code.
- Skipping the self-check, then asking for review too early.
Powstik Guide
A public repo with the computer vision lessons' code and notebooks (convolutions, CNNs, transfer learning, detection, segmentation, vision transformers).
Steps
- Make a public repo for this phase.
- Do one lesson at a time and commit after each.
- Self-check: Run each lesson's code; it must work on your machine as the lesson shows. Each lesson has a quiz (quiz.json, also on the course site): pass it.
- Write a short README: what is here and how to run it.
- Ask three people who know the topic to check it.
Template
README Phase 4: Computer Vision (AI Engineering from Scratch) What is here: How to run it: Self-check result: What was hardest:
What gets sent back
- One big upload with no history.
- Self-check not shown.
- Code that does not run from the README.
Resources
Foundationstart here
Depthgo deeper
What a verifier looks for
- Each lesson's code runs and its quiz is passed.
- Models are evaluated on held-out images.
- Results include the metric the lesson uses (accuracy, mAP, IoU).
- Ask about one choice in the code; the answer should be clear.
You'll sign in first, then come straight back here.
Build an image model for a real use
2–4 weeks (6–10 hrs/week)
Use what you learned. Make an image classifier or detector for a real use you choose (sorting your own photos, spotting defects, counting objects), fine-tuned with transfer learning, with a held-out test set and a confusion matrix. It must run, be evaluated and be documented.
Proof required
Link an image classifier or detector for a real use you choose (sorting your own photos, spotting defects, counting objects), fine-tuned with transfer learning, with a held-out test set and a confusion matrix, with a README that says how to run it and what the evaluation found.
What gets checked
- It runs from the README.
- It is evaluated on images it never saw, with the numbers shown.
- Its worst mistakes are shown and explained.
Common mistakes
- A demo with no evaluation.
- Numbers without the data or code that produced them.
Powstik Guide
An image classifier or detector for a real use you choose (sorting your own photos, spotting defects, counting objects), fine-tuned with transfer learning, with a held-out test set and a confusion matrix.
Steps
- Write one line: what it does and for whom.
- Build the smallest version that works.
- Write the evaluation first, then improve against it.
- Write the README with the results.
- Ask three people to check it.
Template
README What it does: How to run it: How it was evaluated: Results: Limits:
What gets sent back
- Does not run from the README.
- No evaluation.
- Claims more than the evaluation shows.
Resources
Foundationstart here
What a verifier looks for
- It runs from the README.
- It is evaluated on images it never saw, with the numbers shown.
- Its worst mistakes are shown and explained.
- Run it, or watch it run: it must work, be evaluated and be documented.
- Check that the work is the person's own.
You'll sign in first, then come straight back here.