skills.rest
Albatross679
Albatross679COMMUNITY
Explore all skills by this author →
GitHub

ml-workflow

Standardize ML experiment workflows with dataclass configs and W&B tracking.

Updated Mar 13, 2026
Search Tags:#workflow#config#ml#experiments#wandb#rl#sweeps
One-click install
npx skills add https://github.com/Albatross679/0313temp --skill ml-workflow-albatross679
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ml-workflow
Source: https://github.com/Albatross679/0313temp/tree/main/.claude/skills/ml-workflow
Command: npx skills add https://github.com/Albatross679/0313temp --skill ml-workflow-albatross679

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The ML workflow skill captures and enforces your team's conventions for configuring experiments, tracking progress, and coordinating training runs across neural, tree, and RL pipelines.

Core Features & Use Cases

  • Central config system and dataclass-based variants for neural, tree, and RL tasks.
  • Unified experiment tracking and monitoring with W&B, sweep support, and auto-batch/VRAM management.
  • Clear codebase structure guidance, documentation expectations, and scaffolding for new ML projects.
  • Use cases: standardizing experiments, enabling reproducibility, and accelerating collaboration on ML projects.

Quick Start

Create or select a ML workflow config (partN/config.py) and run the corresponding training wrapper to initialize and execute your experiments.

Frequently Asked Questions about ml-workflow

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I standardize ML experiment workflows to reduce setup time and human error?▼

Standardize ML experiment workflows by enforcing a dataclass-based config hierarchy and a documented project structure. This reduces setup time, duplication, and human error across neural, tree, and RL pipelines.

Can I use W&B integration for experiment tracking and sweeps in my ML pipeline?▼

Yes, you can use W&B integration for unified experiment tracking, monitoring, and sweep support. The workflow applies this across neural, tree, and RL pipelines to enable reproducibility and accelerate collaboration.

How do I manage sequential GPU task execution and VRAM with bf16 optimizations?▼

Manage sequential GPU task execution by applying the workflow's enforced VRAM and bf16 optimizations. This standardizes training execution and auto-batch management across your machine learning pipelines.

Does this ML workflow approach support both reinforcement learning and tree-based pipelines?▼

Yes, this ML workflow approach supports reinforcement learning and tree-based pipelines. It provides central config systems with dataclass-based variants specifically designed for neural, tree, and RL tasks.

What is the best way to structure a new ML project for reproducibility and collaboration?▼

The best way to structure a new ML project is using the provided codebase structure guidance and scaffolding. This defines your experiment workflow, establishes documentation expectations, and accelerates team collaboration.

How do I start configuring experiments after defining my ML workflow?▼

To start configuring experiments, create or select an ML workflow config file such as partN/config.py. Run the corresponding training wrapper to initialize and execute your experiments efficiently.

Related Skills

Explore More Software Engineering Skills→
981029l981029l
228

webnovel-init

Initialize a new webnovel project with structured files and directory templates.

Community
Intermediate
granietgraniet
247

weights-and-biases

Log ML experiment metrics, hyperparameters, and artifacts to W&B.

Community
Intermediate
duypham2801duypham2801

serverless-modal

Run GPU workloads on Modal via generated launcher scripts.

Community
Intermediate
DifanaDAPDifanaDAP

weights-and-biases

Automate ML experiment tracking and hyperparameter sweeps via the Python API.

Community
Advanced