Step05
Unlocks soonIntermediateWorkshop In 20 days
Deep Learning Decoded
Neural Networks with PyTorch
- Saturday, 24 October 2026
- 9:30 am – 4:30 pm IST
- Main Seminar Hall, Block A, College of Engineering Kidangoor
Open the hood of deep learning. Build a multilayer network neuron by neuron, then rebuild it in PyTorch, learn to read a training curve like a doctor reads an ECG, and fix the classic failure modes: vanishing gradients, overfitting and learning rates gone wild.
You'll be able to…
- Write a PyTorch training loop from memory
- Diagnose overfitting and underfitting from loss curves
- Use regularisation, dropout and batch norm effectively
- Train on a GPU in Colab
Prerequisites
- Step 03 or equivalent ML basics
What to bring
- Laptop (8 GB RAM recommended) and charger
- Python 3.11+ installed
- A Google account for Colab
- Curiosity — plenty of it
- PyTorch installed or Colab ready
Agenda
- 09:30Check-in & coffeeGrab your badge, find a seat, connect to Wi-Fi.
- 10:00Neurons, layers and the chain rule
- 11:30Short break
- 11:45Guided walkthroughLive coding with the speaker — follow along on your laptop.
- 13:00Lunch
- 14:00Handwritten Malayalam digit recogniserWork in pairs. Mentors float between tables.
- 16:00Show & tellThree teams demo what they built.
- 16:30Wrap-up & next step preview
Speaker
Vishnu Prasad
Deep Learning Engineer, Monsoon AI, Kochi
Vishnu trains large models for a living and debugs exploding gradients for fun. Kaggle Master, PyTorch contributor.
Venue
Main Seminar Hall, Block A, College of Engineering Kidangoor
Kidangoor South P.O., Kottayam, Kerala 686583, India