Lesson 11
Reshape & Flatten
Change structure, keep data:
a = np.arange(12) b = a.reshape(3, 4) b.flatten() # back to 1D
Why this matters
Models and file formats care about shape (batch × features, image × channels). Reshaping lets you reinterpret the same buffer: 1D sequence ↔ grid ↔ batch matrices—often without copying when the total size matches.
Confusing reshape with transpose is a common mistake; this lesson anchors the difference.
ResourcesDocs, references & more — opens in a new tab
🎯 Your Task
flat = np.arange(1,13). Reshape → grid (3,4). Flatten → back_flat.
flat.reshape(3, 4) then grid.flatten()
⌘⏎ run · ⌘← → nav
▶ Output
Run your code to see output here.