I used a Markov chain recently as an example of how to build a plugin for my LLM tool that adds a new model option. Markov chains work quite well as an extremely simple imitation of what language models are doing: https://llm.datasette.io/en/stable/plugins/tutorial-model-plugin.html
Nice! Here is an old hunk of code that looks quite similar, but includes basic end-of-sentence detection and punctuation handling. It’s easy to get started, and the results only get better as the input text gets cleaner.
I miss the King James Programming Bible
At The Modules of Madness is a good one, too.
And, of course, Mark V. Shaney.
I used a Markov chain recently as an example of how to build a plugin for my LLM tool that adds a new model option. Markov chains work quite well as an extremely simple imitation of what language models are doing: https://llm.datasette.io/en/stable/plugins/tutorial-model-plugin.html
Nice! Here is an old hunk of code that looks quite similar, but includes basic end-of-sentence detection and punctuation handling. It’s easy to get started, and the results only get better as the input text gets cleaner.
Great little weekend read. Including the samples 😂 Thanks for posting.