Analyze ML optimization session logs to generate actionable plugin improvement suggestions.
npx skills add https://github.com/ChuaHanChong/ml-optimizer --skill review-chuahanchong
Please help me install this Agent Skill. Skill: review Source: https://github.com/ChuaHanChong/ml-optimizer/tree/main/skills/review Command: npx skills add https://github.com/ChuaHanChong/ml-optimizer --skill review-chuahanchong
This Skill analyzes past optimization runs to identify systemic issues, suggest improvements, and enhance the overall effectiveness and efficiency of the ML optimizer plugin itself.
Use the review skill to analyze the current project's optimization session and identify areas for plugin improvement.
High-intent search queries and answers about installing and using this skill.
Analyzing ML optimization sessions involves processing error logs and experiment outcomes to identify systemic issues and generate actionable plugin improvement suggestions. This skill scopes analysis to single projects or cross-project memory and prioritizes suggestions based on impact and frequency.
Identifying systemic issues from failed ML experiments requires error analysis to detect patterns in agent failures and pipeline inefficiencies. This approach highlights top-performing configurations and effective strategies while isolating recurring problems such as high batch sizes on limited GPU memory.
Generating actionable suggestions for modifying ML optimizer configurations requires analyzing success metrics and proposal outcomes from past optimization runs. This provides specific recommendations for plugin code or configuration modifications, such as adding batch size caps, based on identified failure patterns.
The ML optimization review process requires only the Python standard library. It utilizes specific Python scripts for data loading, pattern detection, success metrics evaluation, proposal outcome analysis, and suggestion logging to process experiment data without external dependencies.
You can analyze cross-project memory to find recurring plugin improvement opportunities by scoping the review to include both single project and cross-project data. This identifies systemic issues and successful strategies across multiple optimization sessions to enhance overall plugin effectiveness.
DMT123Log and visualize ML training metrics via Python API and CLI.
oren-arguAnalyze Trigger.dev task runs and configurations to generate prioritized cost-saving recommendations.
HuskyDannyAutomate MLOps workflows with Weights & Biases for experiment tracking and model versioning.
opendatahub-ioGroup and deduplicate CI/CD pipeline failures using error files and Jira tickets.
