Background
Back to Articles/PyTorch 101: Zero to Hero
Sections
IntroductionModule 01 — Tensors 🧊01 · Creating Tensors 🧊02 · Indexing & Reshaping 🔪03 · Tensor Math ➗04 · Device Placement 🖥️⚡Module 02 — Autograd ⚙️01 · The Computational Graph 🕸️02 · Backward Pass & Gradients ⬅️03 · Turning Autograd Off 🛑04 · Gradient Gotchas 🪤Module 03 — Neural Networks01 · The `nn.Module` Basics 🧱02 · Common Layers 🧩03 · Building a Network 🏗️04 · Inspecting Models 🔍Module 04 — Data Handling 🗂️01 · Dataset Basics 📇02 · The DataLoader 🚚03 · Transforms 🎨04 · Splits & Built-in Datasets ✂️Module 05 — The Training Loop 🔁01 · Loss Functions 🎯02 · Optimizers ⚙️03 · The Training Loop 🔁04 · Evaluation & Metrics 📊Module 06 — Saving & Loading 💾01 · `state_dict` Basics 💾02 · Checkpoints & Resuming ⏸️03 · Loading for Inference 🚀04 · Best Model & Early Stopping 🏅Module 07 — Computer Vision 👁️01 · Convolutions 🔲02 · CNN Architecture 🏛️03 · Transfer Learning 🔁04 · Image Classification Project 🧪Module 08 — NLP & Transformers 💬01 · Text Data & Tokenization 🔤02 · Embeddings 🧭03 · Recurrent Layers & LSTMs 🔄04 · Intro to Transformers ⚡Module 09 — Ecosystem: PyTorch Lightning ⚡01 · Why Lightning? 🤔02 · The LightningModule 🧩03 · The Trainer 🎛️04 · DataModules & Callbacks 🧰Module 10 — Deployment & Optimization 🚀01 · Exporting Models 📦02 · `torch.compile` ⚡03 · Inference Optimization 🪶04 · Serving & Next Steps 🎓

Module 01 — Tensors 🧊

2 min read

The single most important object in PyTorch. Get comfortable here and the rest of your journey gets dramatically easier.

Every image, sentence, and network weight in deep learning is ultimately just a big grid of numbers. The tensor is PyTorch's container for those numbers — a supercharged cousin of the NumPy array that can live on a GPU and remember how it was computed so gradients can flow backward during training. Master the tensor and you've learned the vocabulary the entire rest of PyTorch speaks.

This module is split into four bite-sized sub-modules. Work through them in order.

📚 Sub-Modules

# Sub-Module What you'll learn
01 Creating Tensors Building tensors from data and factory functions, and controlling their dtype
02 Indexing & Reshaping Slicing, indexing, reshape/view, and the crucial views-vs-copies distinction
03 Tensor Math Element-wise ops, matrix multiplication (@), reductions, and broadcasting
04 Device Placement Moving tensors across CPU / CUDA / MPS and bridging to and from NumPy

🎯 By the end of this module, you'll be able to...

  • Create tensors of any shape and data type, from scratch or from existing data.
  • Slice, index, and reshape tensors confidently — without silently corrupting your data.
  • Perform element-wise and matrix math, and know exactly which is which.
  • Move tensors between the CPU and a GPU, and convert to/from NumPy.

✅ Prerequisites

Basic Python and a working PyTorch install. If you haven't set up PyTorch yet, see the Quickstart in the root README.

➡️ Next up: Module 02 — Autograd, where these tensors learn to remember their own history so we can train with them.