isaaccorley/Making-Convolutional-Networks-Shift-Invariant-Again-Tensorflow

Tensorflow Implementation of BlurPool the Antialiasing Pooling operation from "Making Convolutional Networks Shift-Invariant Again"

35
/ 100
Emerging

This project helps machine learning practitioners improve the robustness and reliability of their image classification and analysis models. By integrating a specialized pooling operation, it takes standard image data and produces more stable, accurate model predictions, especially when dealing with slight shifts or movements in the input images. This is for data scientists and AI engineers who build and deploy computer vision systems.

No commits in the last 6 months.

Use this if you are building convolutional neural networks for image processing and want to ensure your models are less sensitive to minor shifts or translations in the images they analyze.

Not ideal if your primary concern is with non-image data or if your existing image processing models already demonstrate sufficient shift invariance.

Computer Vision Image Recognition Deep Learning Machine Learning Engineering AI Model Robustness
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 14 / 25

How are scores calculated?

Stars

10

Forks

3

Language

Python

License

GPL-3.0

Last pushed

Aug 18, 2019

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/isaaccorley/Making-Convolutional-Networks-Shift-Invariant-Again-Tensorflow"

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