Reading Learning Curves: Diagnosing Model Training
Your model finished training. Accuracy: 94%. You lean back, satisfied. Then it meets real data and falls apart like a …
Abhay
4 min read
Your model finished training. Accuracy: 94%. You lean back, satisfied. Then it meets real data and falls apart like a …
Abhay
4 min read
Three little words trip up almost everyone learning deep learning: epoch, iteration, and batch size. They sound like …
Abhay
4 min read
Somewhere along the way, “AI” and “deep learning” became the same word in most people’s …
Abhay
4 min read
A neural network is, at heart, a very confident guesser that is wrong a lot at first. You show it a photo of a cat, it …
Abhay
4 min read
Show a toddler a cat once and they’ll spot cats forever — fluffy ones, grumpy ones, cats half-hidden behind a …
Abhay
4 min read
Imagine stacking ten layers of neurons, training for hours, burning through GPU credits — and ending up with something …
Abhay
4 min read
You already know what a learning rate is: the size of the step your model takes downhill each time it learns. What …
Abhay
4 min read
Every machine learning model, no matter how clever, is fundamentally an over-eager intern with one job: stop being …
Abhay
5 min read
Strip away the hype and almost every machine learning model — from a humble linear regression to a 70-billion-parameter …
Abhay
4 min read
Gradient descent tells you which way is downhill. The optimizer decides how to actually take the step. That distinction …
Abhay
4 min read