PyTorch 101 Crash Course For Beginners in 2026 | Daniel Bourke

Want to master PyTorch? This crash course by ML Engineer Daniel Bourke is the most up-to-date PyTorch tutorial on YouTube!

If you like this, you’ll LOVE Dan's full course! Dive into advanced topics and complete the FoodVision project 👇

🔥 Full PyTorch Bootcamp Course: https://zerotomastery.io/courses/learn-pytorch/
🎁 Use YTPYTORCH10 to get 10% OFF
👨‍💻 Source Code: https://github.com/mrdbourke/pytorch-deep-learning
📓 Course Materials: https://www.learnpytorch.io/
🐍 Free Python 101 Crash Course: https://youtu.be/4uBbCUjJ_G8

==========

⏰ Timestamps ⏰

Ch. 0 – Fundamentals
00:00 Intro
01:34 What is Deep Learning?
07:38 Why Machine Learning?
11:21 Rule of ML
17:11 ML vs. DL
23:27 Neural Network Anatomy
32:58 Learning Paradigms
37:38 DL Applications
44:10 PyTorch Intro
54:32 Tensors
58:57 Course Overview
1:05:13 Best Practices
1:10:32 Resources
1:16:04 PyTorch Setup
1:23:53 Intro to Tensors
1:37:28 Random Tensors
1:47:36 Zeros & Ones
1:50:55 Ranges
1:56:22 Data Types
2:05:56 Attributes
2:14:29 Operations
2:20:38 Matrix Multiplication Pt. 1
2:30:23 Matrix Multiplication Pt. 2
2:38:24 Shape Errors
2:51:31 Aggregation
2:57:51 Min/Max
3:01:17 Reshaping
3:15:08 Squeeze/Unsqueeze
3:27:13 Indexing
3:36:55 Tensors & NumPy
3:46:13 Reproducibility
3:57:10 Accessing GPUs
4:09:10 Device-Agnostic Code
4:17:03 Exercises

Ch. 1 – Workflow
4:22:03 Workflow Intro
4:24:59 Setup
4:32:23 Dataset Creation
4:42:14 Data Splitting
4:50:44 Data Visualization
4:58:39 Linear Model
5:12:59 Model Breakdown
5:19:19 Key PyTorch Classes
5:25:55 Inspect Model
5:35:56 Predictions
5:47:18 Training Intuition
5:55:43 Optimizer Setup
6:08:44 Training Loop
6:22:48 Write Training Loop
6:31:45 Training Steps Review
6:46:52 Run Training Loop
6:56:28 Testing Code
7:08:15 Testing Steps Review
7:23:07 Model Save/Load
7:37:02 Device-Agnostic Practice
7:52:09 Full Workflow: Data
7:58:27 Model
8:08:44 Training
8:21:34 Predictions
8:27:02 Save/Load
8:36:22 Exercises

Ch. 2 – Neural Network Classification
8:40:30 Intro
8:50:21 Example
8:59:38 Architecture
9:06:19 Dataset Creation
9:18:47 Splitting Data
9:30:53 Modelling Steps
9:35:22 Small Network
9:46:29 Visualize Model
9:53:37 Using nn.Sequential
10:07:04 Functions Setup
10:22:05 From Logits to Labels
10:38:21 Training Loops
10:53:58 Predictions
11:08:21 Model Improvement
11:16:34 New Model
11:25:50 Test New Model
11:38:45 Straight Line Dataset
11:47:03 Fit Straight Line
11:57:14 Predictions Evaluation
12:02:47 Adding Non-Linearity
12:12:57 Non-Linear Model
12:23:32 Training Non-Linear Model
12:38:55 Evaluate Non-Linear Model
12:44:52 Activation Functions
12:54:37 Multi-Class Dataset
13:06:11 Multi-Class Model
13:18:49 Multi-Class Loss Function
13:25:39 Logits to Labels (Multi-Class)
13:36:51 Train Multi-Class Model
13:53:18 Evaluate Multi-Class Model
14:01:27 Classification Metrics
14:10:54 Exercises

Ch. 3 – Computer Vision
14:14:02 Intro
14:26:01 Input/Output Shapes
14:36:19 What is CNN?
14:41:32 CV Libraries
14:51:02 Dataset Overview
15:05:43 Visualizing Samples
15:15:45 DataLoader Overview
15:23:13 DataLoaders Creation
15:35:46 Baseline Model
15:50:35 Loss Function & Optimizer
16:01:14 Timing Code
16:06:59 Training & Testing Loops
16:28:35 Evaluation Function
16:41:43 Device-Agnostic Code
16:45:40 Model 1: Non-Linear
16:54:53 Loss Function
16:58:07 Refactor Training Loop
17:06:46 Refactor Testing Loop
17:13:32 Train Model 1
17:25:35 Model 1 Results
17:29:54 Model 2: CNN Overview
17:38:28 Build CNN
17:58:26 Conv2D Explanation
18:13:36 Layer Shapes
18:27:32 CNN Loss Function
18:30:21 Train CNN
18:38:25 Compare Results
18:45:59 Best Model Predictions
18:57:48 Plot Predictions
19:06:09 Confusion Matrix Setup
19:21:40 Evaluate with Confusion Matrix
19:28:45 Save Best Model
19:40:22 Summary/Exercises

Ch. 4 – Custom Datasets
19:46:34 Intro
19:56:38 Device-Agnostic Code
20:02:43 Download Images
20:16:58 Explore Format
20:25:50 Visualize Images
20:37:40 Transform Images
20:42:38 Data Augmentation
21:03:23 ImageFolder Loading
21:12:51 Visualize Loaded Images
21:20:20 DataLoader Creation
21:29:33 Custom Dataset Class
21:37:43 Helper Function
21:47:01 Write Custom Dataset
22:04:57 Dataset Class Comparison
22:12:21 Visualize Custom Dataset
22:26:50 Datasets to DataLoaders
22:33:59 Advanced Augmentation
22:48:33 Baseline Model Overview
22:56:59 Build Tiny VGG
23:08:34 Forward Pass
23:16:54 Torchinfo Summary
23:23:41 Training Functions
23:36:55 Train Model 0
23:47:19 Plot Loss Curves
24:06:36 Overfitting vs. Underfitting
24:20:59 Augmented Datasets
24:32:13 Train Model 1
24:39:34 Compare Loss Curves
24:54:12 Custom Data Predictions
25:38:56 Summary/Exercises

Ch. 5 – Going Modular
25:48:17 Intro
26:00:02 Notebook Pt. 1
26:07:52 Dataset Download
26:12:52 Python Script Outline
26:26:53 PyTorch DataLoaders Script
26:37:40 Model Building Script
26:53:35 Save Model Script
26:59:52 Training Script
27:15:49 Summary/Exercises
27:21:58 Final Takeaway

Full PyTorch Bootcamp 👉 https://zerotomastery.io/courses/learn-pytorch/