skills.rest
marlo9981
marlo9981COMMUNITY
Explore all skills by this author →
GitHub

cuml-machine-learning

Run GPU-accelerated cuML machine learning on tabular datasets.

Updated Mar 17, 2026
Search Tags:#classification#gpu#clustering#machine-learning#rapids#cuml#tabular-data
One-click install
npx skills add https://github.com/marlo9981/Movara --skill cuml-machine-learning-marlo9981
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cuml-machine-learning
Source: https://github.com/marlo9981/Movara/tree/main/examples/nvidia_deep_agent/skills/cuml-machine-learning
Command: npx skills add https://github.com/marlo9981/Movara --skill cuml-machine-learning-marlo9981

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Accelerates common machine learning workflows on tabular data by leveraging NVIDIA RAPIDS cuML to reduce training and inference time for large datasets, while providing a scikit-learn compatible API and sensible CPU fallbacks.

Core Features & Use Cases

  • High-performance training: Train classification and regression models much faster on GPUs for datasets with tens of thousands of rows or more.
  • Unsupervised learning & reduction: Run clustering (KMeans, DBSCAN), dimensionality reduction (PCA, UMAP) and visualization with GPU speedups.
  • Preprocessing & compatibility: Includes guidance for feature scaling, label encoding, and float32 data requirements, and falls back to scikit-learn when GPUs or cuML are unavailable.
  • Use case example: Rapidly prototype a RandomForestClassifier on a large customer dataset to detect churn, report accuracy, and produce ranked feature importances.

Quick Start

Train a GPU-accelerated RandomForestClassifier on data.csv predicting the target column and return dataset shape, train/test sizes, accuracy, and feature importances.

Frequently Asked Questions about cuml-machine-learning

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

FAQPage Schema
How do I accelerate machine learning training on large tabular datasets?▼

GPU-accelerated machine learning on tabular datasets uses RAPIDS cuML to reduce training and inference time for tens of thousands of rows or more, providing a scikit-learn compatible API for faster execution.

Can I run clustering and dimensionality reduction on GPU with scikit-learn compatibility?▼

Yes, GPU-accelerated clustering and dimensionality reduction support KMeans, DBSCAN, PCA, and UMAP workflows using RAPIDS cuML, while maintaining a scikit-learn compatible API and falling back to CPU if no GPU is available.

Do I need an NVIDIA GPU to use RAPIDS cuML for tabular data workflows?▼

An NVIDIA GPU is required for RAPIDS cuML acceleration, but the workflow automatically falls back to scikit-learn on CPU if a GPU or cuML is unavailable, ensuring your tabular data processing continues without interruption.

What data formats are required for GPU-accelerated machine learning with cuML?▼

GPU-accelerated machine learning with cuML requires dense float32 features and avoids sparse matrices. You must convert your tabular datasets to this format before applying classification, regression, or preprocessing workflows.

Why does my GPU machine learning model fall back to scikit-learn on CPU?▼

Your GPU machine learning model falls back to scikit-learn on CPU when NVIDIA GPUs or cuML are unavailable, ensuring compatibility. Check your NVIDIA GPU setup and cuML installation to restore GPU-accelerated training.

What is the best way to train a RandomForestClassifier on large datasets for churn prediction?▼

The best way to train a RandomForestClassifier on large datasets for churn prediction is using GPU-accelerated machine learning with RAPIDS cuML, which rapidly prototypes models to report accuracy and produce ranked feature importances.

Related Skills

Explore More Data & Analytics Skills→
bogwarebogware
3

cuml-machine-learning

Train and evaluate cuML classification, regression, clustering, and dimensionality reduction models on tabular datasets.

Official
Intermediate
HolobiomicsLabHolobiomicsLab
4

scikit-learn

Develop and evaluate scikit-learn machine learning models with pipelines and cross-validation.

Official
Advanced
MrNeo01MrNeo01

cuml-machine-learning

Accelerate tabular machine learning with NVIDIA cuML GPU-accelerated training and inference.

Community
Intermediate
leesk212leesk212
1

cuml-machine-learning

Train and evaluate GPU-accelerated machine learning models on tabular data with cuML.

Community
Intermediate