Deep Neural Networks with PyTorch
summary
A hands-on PyTorch walkthrough showing how to implement neural networks and classic ML models from scratch with runnable notebooks, including a dogs-vs-cats example that shows training curves and GPU speedups. Covers nn.Module model construction, DataLoader with multiple workers, custom Sampler and composed transforms, moving tensors to devices and train/eval mode for dropout/batchnorm, freezing parameters, the optimizer.zero_grad/backward/step loop, basic torch.jit tracing and model export, and an early-stopping demo that halts training when validation stops improving.
AI-written summary · about
chapters (12)
- 0:00PyTorch and deep learning
- 2:09Machine learning core concepts
- 8:22PyTorch setup and imports
- 13:21Tensors and operations
- 23:55GPU acceleration with CUDA
- 35:09Building neural networks
- 40:14Training models and optimization
- 51:40Data loading pipelines
- 1:01:30Samplers and custom data
- 1:10:00Model evaluation and tuning
- 1:17:37Advanced training techniques
- 1:22:16Deploying models
related talks
41mReally Deep Neural Networks with PyTorch
01h 05mDeep Learning with PyTorch
20mLearn You a PyTorch!
01h 40mA Deep Dive into NLP with PyTorch
59mA Friendly Introduction to Deep Learning for Computer Vision
01h 32mIntroduction to PyTorch
32mYou Can Do Deep Learning!
49mDynamic Neural Networks using PyTorch