aiaudit-org/raw2logit

In order to control processing-induced dataset drift we require two ingredients: raw sensor data and an image processing model. This code repository contains the materials for the second ingredient, the image processing models, as well as scripts to load data and run experiments. https://arxiv.org/abs/2211.02578

35
/ 100
Emerging

This project helps machine learning practitioners or researchers who work with optical images from cameras or scientific instruments to understand and control 'dataset drift.' It takes raw sensor data and image processing parameters as input to simulate different image processing pipelines. The output helps analyze how these pipelines affect model performance, enabling more robust and reliable machine learning systems.

No commits in the last 6 months.

Use this if you need to understand how variations in image processing, such as camera settings or instrument calibration, impact the performance of your machine learning models and want to mitigate those risks.

Not ideal if your machine learning models do not use optical images or if you are not concerned with the effects of upstream image processing on your data.

microscopy medical-imaging drone-imagery machine-vision data-quality
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 7 / 25
Maturity 16 / 25
Community 12 / 25

How are scores calculated?

Stars

35

Forks

5

Language

Python

License

MIT

Last pushed

May 07, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/computer-vision/aiaudit-org/raw2logit"

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