MLRichter/receptive_field_analysis_toolbox

A toolbox for receptive field analysis and visualizing neural network architectures

39
/ 100
Emerging

This tool helps machine learning engineers and researchers quickly analyze and optimize convolutional neural network architectures. It takes your existing PyTorch or TensorFlow/Keras model as input and outputs a visual graph that highlights layers predicted to be inefficient for a given input image resolution. This allows you to identify and fix potential issues before costly model training, especially when working with lower resolution images.

116 stars.

Use this if you need to rapidly check and visualize your neural network's architecture for inefficiencies, especially when adapting models for different image resolutions or optimizing performance without extensive training.

Not ideal if you need to analyze models with complex looping logic in their forward pass or if you rely heavily on the functional API for stateful operations in PyTorch.

neural-network-design model-optimization computer-vision deep-learning-research architecture-analysis
No Package No Dependents
Maintenance 6 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 7 / 25

How are scores calculated?

Stars

116

Forks

5

Language

Python

License

MIT

Last pushed

Nov 19, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/MLRichter/receptive_field_analysis_toolbox"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.