niteshade Docs

niteshade (/ˈnaɪt.ʃeɪd/) is a Python library for simulating data poisoning attack and defence strategies against online machine learning systems. The library is written in Python 3.10 and offers a simple and intuitive API which is heavily integrated with PyTorch’s machine learning framework.

For information about data poisoning attacks and why they are a topic of concern as well as general information about the project, check out the About section. For details regarding setting up and using niteshade, see the Installation, Usage and API sections. Are you a cybersecurity researcher or Python developer interested in contributing to niteshade? See the Contributors section.

Contents

• About
  • Adversarial Machine Learning
  • Online Machine Learning
  • Data Poisoning Attacks Against Online Learning
  • Defending Against Data Poisoning Attacks
  • What is niteshade?
  • References
• Installation
• Usage
  • Getting Started
  • Setting Up an Online Data Pipeline
  • Managing Pipeline Asynchronicity
  • Setting Up a Victim Model
  • Defining an Attack Strategy
  • Defining a Defence Strategy
  • Running a Simulation
  • Postprocessing Results
  • End-To-End Example
• API
  • niteshade.data
    • niteshade.data.DataLoader
  • niteshade.models
    • niteshade.models.BaseModel
    • niteshade.models.CifarClassifier
    • niteshade.models.IrisClassifier
    • niteshade.models.MNISTClassifier
  • niteshade.attack
    • niteshade.attack.AddLabeledPointsAttacker
    • niteshade.attack.AddPointsAttacker
    • niteshade.attack.Attacker
    • niteshade.attack.BrewPoison
    • niteshade.attack.ChangeLabelAttacker
    • niteshade.attack.LabelFlipperAttacker
    • niteshade.attack.PerturbPointsAttacker
    • niteshade.attack.RandomAttacker
  • niteshade.defence
    • niteshade.defence.Defender
    • niteshade.defence.DefenderGroup
    • niteshade.defence.Distance_metric
    • niteshade.defence.FeasibleSetDefender
    • niteshade.defence.KNN_Defender
    • niteshade.defence.ModelDefender
    • niteshade.defence.OutlierDefender
    • niteshade.defence.PointModifierDefender
    • niteshade.defence.SoftmaxDefender
  • niteshade.simulation
    • niteshade.simulation.wrap_results
    • niteshade.simulation.Simulator
    • niteshade.simulation.ArgNotFoundError
    • niteshade.simulation.ShapeMismatchError
  • niteshade.postprocessing
    • niteshade.postprocessing.PDF
    • niteshade.postprocessing.PostProcessor
  • niteshade.utils
    • niteshade.utils.check_batch_size
    • niteshade.utils.check_num_of_classes
    • niteshade.utils.copy
    • niteshade.utils.decode_one_hot
    • niteshade.utils.get_cmap
    • niteshade.utils.get_time_stamp_as_string
    • niteshade.utils.load_model
    • niteshade.utils.one_hot_encoding
    • niteshade.utils.rand_cmap
    • niteshade.utils.save_model
    • niteshade.utils.save_pickle
    • niteshade.utils.save_plot
    • niteshade.utils.train_test_MNIST
    • niteshade.utils.train_test_cifar
    • niteshade.utils.train_test_iris
• Contributors