How can I use the resource report in my own scripts?

The skill generates a '.claude_resources.json' file which you can load using Python's json module to dynamically adjust your code's parallelism or memory usage.

What software dependencies are required?

The detection script requires the 'psutil' Python package. All other components use the Python standard library.

Does it detect non-NVIDIA GPUs?

Yes, it supports NVIDIA via nvidia-smi, AMD via rocm-smi, and Apple Silicon via system_profiler, providing a cross-platform hardware overview.

Does this skill support Apple Silicon?

Yes, it specifically detects M1, M2, M3, and M4 chips, identifies unified memory, and recommends Metal-based backends for PyTorch, JAX, and TensorFlow.

Computational Resource Profiler

Name: Computational Resource Profiler
Author: Activer007

byActiver007

Data Science & ML

Analyzes system hardware to provide strategic recommendations for high-performance computing and scientific tasks.

About

The Computational Resource Profiler optimizes resource-intensive workflows by automatically detecting CPU, GPU (NVIDIA, AMD, Apple Silicon), memory, and disk availability to generate actionable strategy recommendations. It assists developers in making informed architectural decisions, such as selecting between parallel processing libraries like Dask or joblib, identifying the correct GPU acceleration backend (CUDA, ROCm, or Metal), and determining whether datasets require out-of-core computing based on available RAM. By providing a structured JSON report of the machine's capabilities, it ensures that scientific scripts and machine learning models are perfectly tuned to their host environment.

Key Features

0 GitHub stars
Standardized JSON output for programmatic integration into CI/CD or data pipelines
Comprehensive GPU detection for NVIDIA CUDA, AMD ROCm, and Apple Silicon Metal
Detailed CPU profiling including architecture, physical cores, and logical cores
Automated strategic recommendations for parallelization and memory management
Real-time analysis of available RAM and disk space with usage snapshots

Use Cases

Managing memory-intensive datasets by switching between Pandas and Dask based on available RAM
Optimizing worker counts for parallel data processing and simulations
Selecting the appropriate deep learning backend before training large models

About

Key Features

0 GitHub stars
Standardized JSON output for programmatic integration into CI/CD or data pipelines
Comprehensive GPU detection for NVIDIA CUDA, AMD ROCm, and Apple Silicon Metal
Detailed CPU profiling including architecture, physical cores, and logical cores
Automated strategic recommendations for parallelization and memory management
Real-time analysis of available RAM and disk space with usage snapshots

Use Cases

Managing memory-intensive datasets by switching between Pandas and Dask based on available RAM
Optimizing worker counts for parallel data processing and simulations
Selecting the appropriate deep learning backend before training large models