- Home
- Courses
- DL Basics — PyTorch MLPs, CNNs, Sequences, Fine-Tuning Discipline
- M1 — Tensors, Autograd + Training Loop
Micro Free Course — 100% Free Learning
All lessons in this module are free to learn. Sign in with Google to save your progress.
M1 — Tensors, Autograd + Training Loop
PyTorch tensors video, MLP loop article, DL-fund quiz, overfit-curve lab submission.
Module Content
PyTorch Tensors + Autograd
Outcome: run gradient check; overfit tiny MLP from scratch. Curated video (13:42, verified 2026-09-24 via yt-dlp): PyTorch Autograd Explained - In-depth Tutorial (Elliot Waite) Watch: https://www.youtube.com/watch?v=MswxJw-8PvE Notes: Autograd basics, backward graphs, complex graph computations, advanced tensor functions. Lab: run a gradient check, then overfit a tiny MLP from scratch (training-loop article follows). Key moments: 0:00 Autograd basics; 1:21 Backward graphs; 3:15 Complex graphs; 5:53 Advanced tensor functions; 8:32 Graph intricacies; 11:09 Finalizing backward passes.
MLP + Training Loop (Loss/Optimizer/Reg)
Diagnose overfit curves (3 cases). Curve-diagnosis drill set.
Quiz: Tensors + Training
10 scenario MCQs. Pass >= 80.
Overfit-Curve Diagnosis Lab
Diagnosis + fix (reg/LR/augment). Curve bundle + 100-pt rubric inside.