train-test-splitter

Train Test Splitter - Auto-activating skill for ML Training. Triggers on: train test splitter, train test splitter Part of the ML Training skill category.

Install with Tessl CLI

npx tessl i github:jeremylongshore/claude-code-plugins-plus-skills --skill train-test-splitter

What are skills?

1.04x

Quality

Does it follow best practices?

Impact

89%

1.04x

Average score across 3 eval scenarios

Optimize this skill with Tessl

npx tessl skill review --optimize ./planned-skills/generated/07-ml-training/train-test-splitter/SKILL.md

SKILL.md

Review

Evals

Evaluation results

90%

Customer Churn Prediction: Data Preparation Pipeline

Stratified classification split with sklearn

Criteria

Without context

With context

Uses sklearn

100%

Stratified split

100%

Random seed set

100%

No preprocessing leakage

100%

Split output files

100%

Python implementation

100%

Split summary printed

100%

pip dependencies

Production code quality

100%

Without context: $0.4378 · 1m 48s · 22 turns · 23 in / 6,216 out tokens

With context: $0.5157 · 2m 6s · 27 turns · 27 in / 6,614 out tokens

78%

-7%

Image Classification: Building a Deep Learning Data Pipeline

Deep learning data pipeline with train/val/test split

Criteria

Without context

With context

PyTorch or TensorFlow used

100%

Three-way split

100%

Train shuffle, test no shuffle

100%

Preprocessing after split

66%

53%

Data loader created

100%

Batch demo output

100%

Pipeline report produced

100%

pip installation

Reproducible seed

100%

50%

Without context: $0.4472 · 2m 31s · 25 turns · 25 in / 7,561 out tokens

With context: $0.7921 · 5m · 37 turns · 298 in / 11,423 out tokens

100%

20%

Predictive Maintenance: Model Selection with Hyperparameter Tuning

Hyperparameter tuning with experiment tracking

Criteria

Without context

With context

sklearn or supported framework

100%

Three-way split enforced

100%

Test set isolation

100%

Multiple hyperparameter configs

100%

Experiment log written

100%

Test score in log

100%

Reproducible seed

100%

Python script produced

100%

pip dependencies

100%

Without context: $0.5067 · 2m 6s · 23 turns · 23 in / 8,620 out tokens

With context: $0.7138 · 2m 55s · 33 turns · 93 in / 10,445 out tokens

Evaluated: 6 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Customer Churn Prediction: Data Preparation Pipeline Image Classification: Building a Deep Learning Data Pipeline Predictive Maintenance: Model Selection with Hyperparameter Tuning

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.