Opening PyTorch Tutorial
PyTorch Tutorial
- Length
- 29 min
- Level
- Undergraduate
- Recorded
- MIT2026
Chapter 1 of 5: Before the tutorial
PyTorch tensors and automatic differentiation form the base of an introductory recitation for MIT 6.7960: Deep Learning, taught by instructor Jamie Meindl. Meindl starts with tensor creation from Python lists, NumPy arrays and fixed shapes, then covers indexing, slicing and element-wise arithmetic. Dimension handling follows, with permute to reorder axes and view or reshape to change shape. The autograd engine records computational history, and a backward pass computes the gradients that SGD and Adam use to update weights. Linear layers and activations, assembled in a sequential container, then classify points relative to a sine wave. Python, NumPy, basic linear algebra and the course's earlier material on gradients are assumed.

Your comment is kept while you sign up.Sign in
Builds a PyTorch network from raw tensors and autograd, then trains it to split points around a sine wave.
PyTorch tensors and automatic differentiation form the base of an introductory recitation for MIT 6.7960: Deep Learning, taught by instructor Jamie Meindl. Meindl starts with tensor creation from Python lists, NumPy arrays and fixed shapes, then covers indexing, slicing and element-wise arithmetic. Dimension handling follows, with permute to reorder axes and view or reshape to change shape. The autograd engine records computational history, and a backward pass computes the gradients that SGD and Adam use to update weights. Linear layers and activations, assembled in a sequential container, then classify points relative to a sine wave. Python, NumPy, basic linear algebra and the course's earlier material on gradients are assumed.
This lecture
Lecture 24 of 24 of Deep Learning.
Choose a node to see how it connects to this lecture. What comes before is on the leftat the top. The ring is what it teaches. People and places are above. Other lectures and books are below.
Press a chosen node again to open it.