Welcome to neuralde’s documentation!

neuralde

Presentation

🔍 Identity card

• Version: 1.1.0

• Python Version: 3.9, 3.10

• Strong Dependencies: Tensorflow, pytorch

• Trustworthy attributes : Robustness

• Hardware : CPU / GPU

• Engineering activities : Data Engineer, ML-Algorithm Engineer

• Functional set : Robustness, Operation, Evaluation

General presentation

Neuralde is a python library made to improve the robustness of your models at test time. It proposes a set of methods that will allow you to remove identified corruptions in your data before you send it to your model. This library addresses issues such as meteorological corruptions, distribution shifts etc.

All methods provided by the library are associated with examples based on open sourced images. You can refer to each jupyter notebook for specifics. These notebooks are linked directly in the technical documentation.

This library can be used in two ways:

• Call it as a main executable to direct process a batch of input images.

• Import the lib within your python code for an explicit call of included transformations functions

Contents:

• 📖 Guidelines
  • Getting started
    • Set up a clean virtual environnement
    • Install the library (restricted access)
    • Use the executable
    • Use the docker version
  • Define your configuration pipeline
    • Available methods in release 1.1.0
  • Required Hardware
  • Usage of the lib inside your python code
    • Detailed examples with notebooks
  • Additional informations
• 📚 Technical docs
• 💡 Theory Overview
• 🔄 Changelog
  • v 1.1.0 (2025-03-31)
  • v 1.0.0 (2024-01-31)
• 🎡 Package neuralde
  • neural_de package
    • Subpackages
    • Submodules
    • neural_de.main module
    • Module contents

Indices and tables

• Index

• Module Index

• Search Page