Graph ML Course
Graph Neural Networks Part 2 (Implementation Included)
A practical and beginner-friendly guide to building neural networks on graph data.
Avi Chawla
367 posts published
Graph ML Course
Graph Neural Networks Part 1 (Implementation Included)
A practical and beginner-friendly guide to building neural networks on graph data.
Classical ML and Deep Learning
Model Calibration (Part 2)
How to make ML models reflect true probabilities in their predictions?
Classical ML and Deep Learning
Federated Learning: A Critical Step Towards Privacy-Preserving ML
Learn real-world ML model development with a primary focus on data privacy – A practical guide.
11 Types of Variables in a Dataset
Beyond features, targets, etc.
Classical ML and Deep Learning
Model Calibration (Part 1)
How to make ML models reflect true probabilities in their predictions?
Engineering Best Practices
A Practical Guide to Scaling ML Model Training
GPUs - GPU Clusters - Distributed Training.
Deep Learning
All-Reduce and Ring-Reduce for Model Synchronization in Multi-GPU Training
Two synchronization algorithms for intermediate-ML models.
Classical ML and Deep Learning
Conformal Predictions: Build Confidence in Your ML Model's Predictions
A critical step towards building and using ML models reliably.
Engineering Best Practices
Quantization: Optimize ML Models to Run Them on Tiny Hardware
A must-know skill for ML engineers to reduce model footprint and inference time.
Engineering Best Practices
How to Structure Your Code for Machine Learning Development
A highly overlooked yet critical skill for data scientists.
Engineering Best Practices
Develop an Elegant Testing Framework For Data Science Projects Using Pytest
A Comprehensive Guide to Pytest for data scientists.
Engineering Best Practices
A Beginner-friendly Guide to Multi-GPU Model Training
Models are becoming bigger and bigger. Learn how to scale models using distributed training.
Engineering Best Practices
Object-Oriented Programming with Python for Data Scientists
A beginner to advanced guide for Python OOP.
Engineering Best Practices
11 Powerful Techniques To Supercharge Your ML Models
Take your ML models to the next level with 11 lesser-known techniques.
The Mathematics Behind RBF Kernel
...and why it so powerful
Machine Learning
Shuffle Feature Importance
An intuitive and reliable technique to measure feature importance.
31 min read
Model Compression: A Step Towards Efficient Machine Learning
Model accuracy alone (or an equivalent performance metric) rarely determines which model will be deployed. Much of the engineering effort goes into making the model production-friendly. Because typically, the model that gets shipped is NEVER solely determined by performance — a misconception that many have. Instead, we also consider several operational
26 min read
Federated Learning: A Critical Step Towards Privacy-Preserving Machine Learning.
There’s so much data on your mobile phone right now — images, text messages, etc. And this is just about one user — you. But applications can have millions of users. The amount of data we can train ML models on is unfathomable. The problem? This data is private. So you
15 min read
A Beginner-friendly Guide to Multi-GPU Training
If you look at job descriptions for Applied ML or ML engineer roles on LinkedIn, most of them demand skills like the ability to train models on large datasets: Of course, this is not something new or emerging. But the reason they explicitly mention “large datasets” is quite simple to
17 min read
5 Must-Know Ways to Test ML Models in Production (Implementation Included)
Despite rigorously testing an ML model locally (on validation and test sets), it could be a terrible idea to instantly replace the previous model with the new model. A more reliable strategy is to test the model in production (yes, on real-world incoming data). While this might sound risky, ML
20 min read
Bayesian Optimization for Hyperparameter Tuning
There are many issues with grid search and random search. * They are computationally expensive due to exhaustive search. * The search is restricted to the specified hyperparameter range. But what if the ideal hyperparameter exists outside that range? * They can ONLY perform discrete searches, even if the hyperparameter is continuous. Bayesian
Classical ML and Deep Learning
Implementing KANs From Scratch Using PyTorch
A step-by-step demonstration of an emerging neural network architecture — KANs.
Classical ML and Deep Learning
A Beginner-friendly Introduction to Kolmogorov Arnold Networks (KAN)
What are KANs, how are they trained, and what makes them so powerful?
Engineering Best Practices
Implementing Parallelized CUDA Programs From Scratch Using CUDA Programming
A beginner-friendly guide for curious minds who don't know the internal workings of model.cuda().
Engineering Best Practices
8 Fatal (Yet Non-obvious) Pitfalls and Cautionary Measures in Data Science
8 data science lesson I wish I had known earlier.
Classical ML and Deep Learning
15 Ways to Optimize Neural Network Training (With Implementation)
Techniques that help you become a "machine learning engineer" from a "machine learning model developer."
Classical ML and Deep Learning
A Detailed and Beginner-Friendly Introduction to PyTorch Lightning: The Supercharged PyTorch
Immensely simplify deep learning model building with PyTorch Lightning.
LLM and Fine-tuning
Augmenting LLMs: Fine-Tuning or RAG?
Understanding the tradeoffs between RAG and Fine-tuning, and owning the model vs. using a third-party host.
LLM and Fine-tuning
Implementing DoRA (an Improved LoRA) from Scratch
A powerful LoRA-variant explained in a beginner-friendly way and implemented in PyTorch.
