Sigmoid and softmax are both functions that turn model scores into values that are easier to interpret, but they solve different kinds of classification problems...
Continue reading → Tags : ml python
A single local Canton sandbox is good for fast contract development. It gives you one participant, one Ledger API, one JSON API, and a simple place to upload DAR files. But it does not help much when you want to reason about distributed behavior: multiple participants, party hosting, package vetting on more than one participant, and the role of a synchronizer...
Continue reading → Tags : canton daml blockchain
These are my learning notes and reflections while reading "AI Engineering: Building Applications with Foundation Models" by Chip Huyen. I plan to update this periodically as I progress through the book...
Continue reading → Tags : books ai llm