skills.rest
dotlab-hq
dotlab-hqOFFICIAL
Explore all skills by this author →
GitHub

cudf-analytics

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

Updated Apr 12, 2026
Search Tags:#profiling#data-analysis#gpu#anomaly-detection#groupby#rapids#cudf
One-click install
npx skills add https://github.com/dotlab-hq/torque --skill cudf-analytics-dotlab-hq
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cudf-analytics
Source: https://github.com/dotlab-hq/torque/tree/main/.agents/skills/cudf-analytics
Command: npx skills add https://github.com/dotlab-hq/torque --skill cudf-analytics-dotlab-hq

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill removes the bottleneck of CPU-bound tabular analysis by enabling high-performance, GPU-accelerated operations on large CSVs and dataframes so users can get fast summaries, aggregations, and anomaly detection on millions of rows.

Core Features & Use Cases

  • Pandas-like GPU computations: Use NVIDIA cuDF with a familiar API to accelerate common pandas workflows.
  • Statistical summaries & profiling: Compute mean, median, std, quantiles, and correlation matrices at scale.
  • Groupby aggregations & reporting: Fast groupby/aggregate operations and conversion to pandas for downstream reporting.
  • Anomaly detection: Built-in examples for IQR and Z-score based outlier detection for data quality checks.
  • Real-world use: Process multi-million-row sales or telemetry CSVs to produce per-category aggregates, detect outliers, and export summary tables for BI dashboards.

Quick Start

Use the cudf-analytics skill to read a large CSV, compute groupby revenue aggregates by category, produce summary statistics, and return the results as a pandas DataFrame.

Frequently Asked Questions about cudf-analytics

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

FAQPage Schema
How do I accelerate pandas groupby aggregations on a large CSV file?▼

GPU-accelerated data analysis with cuDF uses a pandas-compatible API to process large tabular datasets. It computes statistical summaries, correlation matrices, and groupby aggregations directly on the GPU to bypass CPU bottlenecks.

Can I use a GPU to detect anomalies and outliers in a large dataframe?▼

Yes, you can use GPU-accelerated cuDF to detect anomalies in large dataframes using built-in IQR and Z-score methods. This allows fast outlier identification across millions of rows for data quality checks.

Do I need an NVIDIA GPU to run GPU-accelerated data profiling on large datasets?▼

Yes, you need an NVIDIA GPU-backed cuDF runtime to perform GPU-accelerated data profiling. The Skill provides fallbacks to pandas and Arrow for host-side conversion and reporting when GPU execution is unavailable.

What is the best way to compute statistical summaries for multi-million-row tabular datasets?▼

The best way to compute statistical summaries for multi-million-row datasets is using GPU-accelerated cuDF. It rapidly calculates mean, median, standard deviation, quantiles, and correlation matrices at scale, returning results as pandas DataFrames.

How does GPU-accelerated dataframe processing compare to standard pandas for large scale analysis?▼

GPU-accelerated dataframe processing removes the CPU bottleneck of standard pandas by utilizing NVIDIA cuDF for large scale analysis. It maintains a familiar pandas-compatible API while providing fallbacks to pandas and Arrow for downstream host-side conversion.

What are the limitations of using cuDF for tabular data analysis?▼

The primary limitation of using cuDF for tabular data analysis is the strict requirement for an NVIDIA GPU-backed runtime. Host-side processing relies on pandas and Arrow fallbacks for conversion and reporting rather than GPU acceleration.

Related Skills

Explore More Data & Analytics Skills→
bogwarebogware
3

cudf-analytics

Accelerate statistical profiling and anomaly detection on tabular data with NVIDIA cuDF.

Official
Intermediate
anukkrit149anukkrit149

cudf-analytics

Profile tabular datasets with NVIDIA cuDF and fallback to pandas.

Community
Intermediate
eren726290eren726290
37

statistical-analysis

Apply descriptive statistics, trend analysis, outlier detection, and hypothesis testing to data using Python.

Community
Advanced
leesk212leesk212
1

cudf-analytics

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

Community
Advanced