---
title: "What model architecture means"
lesson_id: "01"
---

# What model architecture means

- **Lesson ID:** 01
- **Goal:** Explain architecture as a model input, output, and information-moving shape. Keep it separate from learned weights.
- **Human lesson:** [01-what-architecture-means.html](01-what-architecture-means.html)

## Prerequisites

- Familiarity with a trained model mapping an input to an output.
- No prior neural-network math.

## Inputs, outputs, and artifacts

- **Inputs:** The reader's understanding of an app feature and the two linked published courses.
- **Outputs:** A plain distinction between architecture and weights.
- **Artifacts:** Three progressive `.flow` diagrams in the human lesson.

## Agent build steps

1. Define architecture as the model shape. Name its input, information path, and output.
2. Define weights as learned settings that fill that shape.
3. Use ClothingCNN from [the forward course](../pytorch-core-ai-xcode/) as one image-classifier architecture choice.
4. Use [From the metal to the model](../core-ai-from-metal-back/) to connect different model shapes to runtime work on silicon.
5. Check that the lesson moves from general shape to ClothingCNN and then to hardware.

## Constraints

Keep the language simple, for someone who is comfortable with computers. Use input-to-output jobs. Do not add math deep-dives, papers, CUDA, or quizzes. Keep both prior-course links relative. Do not use em dashes. Keep diagrams as HTML `.flow` elements, not Mermaid.

## Key concepts

- Architecture is the layout: what the model reads, how information moves, what it writes.
- Weights are the learned settings inside that layout.
- ClothingCNN is an image-classifier shape: 28x28 clothing pixels in, a clothing label out.
- Different shapes still become work on the chip when the app runs.

## Takeaways

- You can say architecture versus weights in one sentence.
- You can describe ClothingCNN as pixels, then model shape, then label.
- You can point to the forward course for the build and From the metal to the model for the hardware path.

## Acceptance checks

- The page separates architecture from weights in plain language.
- ClothingCNN is clearly identified as an image-classifier architecture choice.
- Both prior courses are linked with relative URLs.
- The page contains progressive `.flow` diagrams and an Agent brief link.
- [ ] Proceed to [lesson 02 brief](02-language-models.llms.md).