wassname/ml-debug
1
0
Fork 0
mirror of https://github.com/wassname/ml-debug.git synced 2026-06-27 16:46:04 +08:00
Code Issues Packages Projects Releases Wiki Activity

name

Browse Source
This commit is contained in:
wassname
2026-04-09 05:09:25 +08:00
parent b159b0fba8
commit fde5ac62fd
3 changed files with 3 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files
SKILL.md
+1 -1
View File
@@ -1,5 +1,5 @@
--- ---
name: ml_debug name: ml-debug
description: "Wassname's practical folklore for debugging ML systems: convergence issues, loss surface analysis, gradient analysis, sweep methodology, and same-seed comparisons. Use when stuck on training, designing sweeps, or analyzing experiment results." description: "Wassname's practical folklore for debugging ML systems: convergence issues, loss surface analysis, gradient analysis, sweep methodology, and same-seed comparisons. Use when stuck on training, designing sweeps, or analyzing experiment results."
--- ---
pinn/SKILL.md
+1 -1
View File
@@ -1,5 +1,5 @@
--- ---
name: pinn-debug name: pinn
description: "PINN (Physics-Informed Neural Network) training best practices and debugging. Use when building, debugging, or optimizing PINNs for PDEs, ODEs, or physics-constrained learning problems. Sub-skill of ml-debugging." description: "PINN (Physics-Informed Neural Network) training best practices and debugging. Use when building, debugging, or optimizing PINNs for PDEs, ODEs, or physics-constrained learning problems. Sub-skill of ml-debugging."
--- ---
rl/SKILL.md
+1 -1
View File
@@ -1,5 +1,5 @@
--- ---
name: rl-debug name: rl
description: "RL-specific debugging: probe environments, reward engineering, diagnostics, hyperparameter defaults, and reference implementations. Sub-skill of ml-debugging. Use when debugging reinforcement learning systems." description: "RL-specific debugging: probe environments, reward engineering, diagnostics, hyperparameter defaults, and reference implementations. Sub-skill of ml-debugging. Use when debugging reinforcement learning systems."
--- ---