From 6e149bde5edd66061b7bec7ec69f1474f5452933 Mon Sep 17 00:00:00 2001 From: wassname <1103714+wassname@users.noreply.github.com> Date: Thu, 27 Aug 2026 09:45:19 +0800 Subject: [PATCH] put be-diligent at the top, it is the part with measured uplift --- SKILL.md | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-) diff --git a/SKILL.md b/SKILL.md index 32a3a69..e35f2e4 100644 --- a/SKILL.md +++ b/SKILL.md @@ -5,7 +5,14 @@ description: "Machine learning debugging exercises, each under a quote from a pr In an attempt to upskill the machine learning debugging on AI coding assistants (and humans), I've collected high quality sources on how to debug machine learning projects, focusing on the mindset and the "taste". When I started ML I went searching for discussions on best practices, and started a few discussions of my own and they helped me a lot, over the years I've collected good ones. I hope they can help others, as well as help in auto research setups. This intro is human written, and the below is AI written with human guidance. - wassname -If this skill loaded, do the exercises for your situation and show the result in your reply. Each +Be diligent. Work the problem in full before you write. State the decisive point early, then give +the derivation, the mechanism, or the log line behind it, so the reader can check it and not just +take it. Show the work, not only the conclusion. + + +Then do the exercises for your situation and show the result in your reply. Each exercise is marked (small) or (large). A small one takes less than a paragraph to answer; a large one is a lot of work, such as searching for a reference repo and comparing against it.