Notes on machine learning
Technical notes written while learning: honest evaluation, data leakage, deployment, LLM applications and the mistakes behind each lesson. They come from the projects on this site, with the code one click away.
Calibrated probabilities: when a 0.7 has to mean 70%
A classifier can rank well and still lie with its numbers: saying 0.9 where the real frequency is 0.6. If someone is going to decide with that probability, you calibrate it, check it with the reliability curve and the Brier score, and only then pick a threshold.
Read article
What I learned implementing a transformer from scratch
Multi-head attention, causal masking and a BPE tokenizer handwritten in PyTorch, without torch.nn.Transformer. Three lessons no course had given me, including a surprise in the ablations.
Read article
Why the decision threshold matters more than the model
Moving the decision threshold from 0.5 to the right value for your business cost can improve the operational result more than switching from logistic regression to XGBoost. Why nobody explains it that way.
Read articleShall we connect?
Let's talk: feedback, collaboration or an opportunity.
You can write to me about a project, a technical question, to give me feedback, or about an internship or a first junior role. I always reply.
Write to me