Shapes and broadcasting on the How to Make AI Models track. Most neural net bugs are shape bugs. Write the shape beside every tensor in a forward pass. Batch dimension first is the usual convention.
This lesson assumes you already worked through Writing it versus using a library.
The idea in practice
Assert shapes in the forward method while you learn. Remove only noisy asserts later.
A concrete check
goal = {
'track': 'How to Make AI Models',
'lesson': 'Shapes and broadcasting',
}
checks = [
'input available at decision time',
'score matches the real decision',
'failure case written down',
]
print(goal['lesson'])
for item in checks:
print('-', item)
Run the sketch locally if you have Python. The printout is a reminder of the checks, not a trained model. Replace the strings with the real inputs from your own example before you treat it as a design.
What usually goes wrong
A silent broadcast that pairs the wrong rows. When this happens, stop adding parameters or tools. Fix the check, the data, or the permission, then run the same example again.
What to write down
- The input you are allowed to use at decision time.
- The output and the score or pass rule.
- One failure you will test on purpose.
- What you will not claim the system can do.
Practice
Annotate shapes for input (32, 20) through a linear layer to 4 classes.
Self-check
- Say Shapes and broadcasting in one sentence that mentions an input and an output.
- Name the failure mode in this lesson and the check that would catch it.
Done when: you can explain this lesson without the page open, and you have a written failure case.