skills.rest
najicham
najichamCOMMUNITY
Explore all skills by this author →
GitHub

model-experiment

Train and evaluate NBA player prop regression and breakout classifier models.

1|Updated May 24, 2025
Search Tags:#evaluation#train#model-experiments#CatBoost#ml-retraining#nba-props
One-click install
npx skills add https://github.com/najicham/nba-stats-scraper --skill model-experiment
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: model-experiment
Source: https://github.com/najicham/nba-stats-scraper/tree/main/.claude/skills/model-experiment
Command: npx skills add https://github.com/najicham/nba-stats-scraper --skill model-experiment

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill enables teams to train and evaluate challenger machine learning models for NBA player prop predictions, enabling rapid experimentation against baselines.

Core Features & Use Cases

  • Train regression models on recent data and compare to V9 baseline.
  • Train breakout classifier models to identify breakout games.
  • Support monthly retraining workflows and easy experiment tracking.

Quick Start

Use the model-experiment skill to kick off common experiments:

  • Default regression retrain: PYTHONPATH=. python ml/experiments/quick_retrain.py --name "FEB_MONTHLY"
  • Breakout classifier training: PYTHONPATH=. python ml/experiments/train_breakout_classifier.py --name "BREAKOUT_V1"
  • Dry run: PYTHONPATH=. python ml/experiments/quick_retrain.py --name "TEST" --dry-run

Frequently Asked Questions about model-experiment

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

FAQPage Schema
How do I retrain NBA player prop models using CatBoost?▼

Challenger model evaluation lets you train new models on recent data and compare their performance against a baseline to determine if they should replace the current production model.

How do I train a breakout classifier for NBA props predictions?▼

Run a dry test of Challenger model training by appending the --dry-run flag to the quick_retrain.py script command, validating the workflow without consuming compute resources.

Can I compare regression and breakout classifier models in the same workflow?▼

You need a Python environment with CatBoost and the related ml experiment scripts installed to run the model training workflows, requiring no other external dependencies.

What is challenger model evaluation in machine learning?▼

Challenger model evaluation lets you train new models on recent data and compare their performance against a baseline to determine if they should replace the current production model.

How do I run a dry test of challenger model training?▼

Run a dry test of Challenger model training by appending the --dry-run flag to the quick_retrain.py script command, validating the workflow without consuming compute resources.

Related Skills

Explore More Data & Analytics Skills→
ChuaHanChongChuaHanChong

baseline

Run evaluation commands and profile GPU usage to establish ML model baseline metrics.

Community
Advanced
rongarederongarede
4

eval-harness

Evaluate Claude Code sessions with capability and regression evals.

Community
Advanced
enigmaicon-engenigmaicon-eng
1

ml-pipeline

Generate production ML training pipelines with MLflow logging and Kubeflow or Airflow orchestration.

Community
Advanced
snoodleboot-iosnoodleboot-io
2

ensemble-methods

Combine predictions from multiple models to construct ensemble models.

Official
Intermediate