WORKSHOP SESSIONS 🧠
These notebooks are designed to be browsed directly on this site. You can also work with them interactively using the "Code Along" button or check the solutions with the "Solution" button. For each session, you can also access the presentation slides using the "Presentation" button.
Note: When using Colab, make sure to enable GPU acceleration for best performance. Go to Runtime → Change runtime type and select GPU as the Hardware accelerator.
SESSION 1: INTRODUCTION TO PYTORCH
This notebook covers the fundamentals of PyTorch, including:
- Tensor operations and manipulation
- Autograd and computational graphs
- Basic GPU operations
Code Along
SESSION 2: ARTIFICIAL NEURAL NETWORKS
Learn how to build neural networks from scratch with PyTorch:
- Neural network architecture
- Forward and backward propagation
- Building custom neural network modules
SESSION 3: MODEL TRAINING & OPTIMIZATION
Learn techniques for effective model training:
- Optimization algorithms
- Regularization techniques
- Performance monitoring
SESSION 3B: PHYSICS-INFORMED NEURAL NETWORKS
Explore how to integrate physics constraints into neural networks:
- Physics-informed loss functions
- Boundary conditions in neural networks
- Solving differential equations with PINNs
SESSION 4: CONVOLUTIONAL NEURAL NETWORKS
Dive into image processing with CNNs:
- Understanding convolutional layers
- Building CNN architectures
- Image classification tasks
SESSION 5: TRANSFER LEARNING & U-NET
Leverage pre-trained models for your tasks:
- Using pre-trained models
- Fine-tuning strategies
- Image segmentation with U-Net
USING THE EXERCISE CHECKER ✅
Throughout the notebooks, you'll find 🎯 exercises. Use the built-in checker to validate your answers.
answer = {'your_solution': result}
checker.check_exercise(1, answer)REQUESTING HINTS 💡
checker.display_hints(1)✔️ Correct = green check
❌ Incorrect = feedback provided
💬 Hints are tailored to the task