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

cudf-analytics

Analyze large CSV and parquet datasets with GPU-accelerated cuDF aggregations and anomaly detection.

Updated Mar 17, 2026
Search Tags:#profiling#data-analysis#gpu#anomaly-detection#groupby#rapids#cudf
One-click install
npx skills add https://github.com/marlo9981/Movara --skill cudf-analytics-marlo9981
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cudf-analytics
Source: https://github.com/marlo9981/Movara/tree/main/examples/nvidia_deep_agent/skills/cudf-analytics
Command: npx skills add https://github.com/marlo9981/Movara --skill cudf-analytics-marlo9981

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Enables fast, GPU-accelerated analysis of large tabular datasets that would be slow or infeasible on CPU-bound pandas workflows, reducing analysis time and enabling interactive exploration of millions of rows.

Core Features & Use Cases

  • High-performance read and processing for CSV and parquet datasets using NVIDIA RAPIDS cuDF with a pandas-like API.
  • Statistical summaries & aggregations including groupby operations, correlation matrices, and dataset profiling at scale.
  • Anomaly detection & data quality checks via IQR and z-score methods, plus guidance for dtype best practices and fallback to pandas when GPUs are unavailable.
  • Use Case: Profile a 50M-row sales dataset to compute per-region revenue aggregations, highlight outliers, and export summarized results for reporting.

Quick Start

Analyze data.csv using GPU-accelerated groupby aggregations, return summary statistics, and list detected anomalies.

Frequently Asked Questions about cudf-analytics

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

FAQPage Schema
How do I perform GPU-accelerated groupby aggregations on a large CSV file?▼

You can perform GPU-accelerated groupby aggregations on large CSV files using NVIDIA RAPIDS cuDF, which provides a pandas-like API to process millions of rows efficiently and compute statistical summaries.

What is the best way to detect anomalies in a large parquet dataset?▼

Anomaly detection in large parquet datasets is handled using IQR and z-score methods to highlight outliers, running on NVIDIA RAPIDS cuDF for GPU acceleration to process millions of rows quickly.

Do I need a GPU to run data profiling on millions of rows?▼

A GPU is not strictly required to run data profiling on millions of rows because the workflow uses NVIDIA RAPIDS cuDF with an automatic fallback to pandas when GPUs are unavailable.

How does cuDF compare to pandas for computing correlation matrices?▼

cuDF provides a pandas-like API for computing correlation matrices but uses GPU acceleration to process large tabular datasets significantly faster than CPU-bound pandas workflows.

Can I use this for statistical summaries on a 50M-row sales dataset?▼

Yes, you can compute statistical summaries and per-region revenue aggregations on a 50M-row sales dataset, leveraging GPU acceleration to enable interactive exploration and export summarized results.

Related Skills

Explore More Data & Analytics Skills→
coinvest518coinvest518

cudf-analytics

Analyze large tabular datasets with NVIDIA cuDF for GPU-accelerated grouping, summarization, and anomaly detection.

Community
Intermediate
dotlab-hqdotlab-hq

cudf-analytics

Perform GPU-accelerated groupby aggregations and statistical summaries on large CSV datasets.

Official
Intermediate
ovachieverovachiever
52

nemo-curator

Curate multimodal LLM training data with GPU-accelerated filtering and deduplication.

Community
Advanced
leesk212leesk212
1

cudf-analytics

Performs GPU-accelerated groupby, descriptive statistics, and anomaly detection on tabular data with cuDF.

Community
Advanced