cnn-families-and-selection

Community

Select the optimal CNN for vision tasks, fast.

Authortachyon-beep
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill guides you through the diverse landscape of CNN architectures (ResNet, EfficientNet, MobileNet, etc.), helping you select the perfect model for your computer vision task based on critical constraints like deployment target, latency, model size, and dataset size. It prevents you from using inefficient or unsuitable models, saving compute and development time.

Core Features & Use Cases

  • Constraint-Based Selection: Get tailored recommendations for cloud, edge, or mobile deployment, considering real-time latency and parameter budgets.
  • Efficiency vs. Accuracy Trade-offs: Understand the Pareto frontier of CNNs, choosing models like EfficientNet for best accuracy per FLOP or MobileNet for extreme mobile efficiency.
  • Use Case: You need to deploy an image classifier on a mobile device with strict latency requirements. This skill directs you to MobileNetV3-Small, emphasizing INT8 quantization for optimal on-device performance.

Quick Start

I need to classify images on an edge device with a latency target of 50ms. Which CNN should I use?

Dependency Matrix

Required Modules

torchvisiontimm

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: cnn-families-and-selection
Download link: https://github.com/tachyon-beep/skillpacks/archive/main.zip#cnn-families-and-selection

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 223,000+ vetted skills library on demand.