v0.1.3

About

Relying on TensorFlow and Keras as backends, pidgan is a Python package designed to simplify the implementation and training of GAN-based models intended for High Energy Physics (HEP) applications.

Available modules

• algorithms
  • BceGAN - 🧩
  • BceGAN_ALP
  • BceGAN_GP - 🧩
  • CramerGAN - 🧩
  • GAN - 🧩
  • LSGAN - 🧩
  • WGAN - 🧩
  • WGAN_ALP
  • WGAN_GP - 🧩
• callbacks
  • schedulers
    • LearnRateBaseScheduler
    • LearnRateCosineDecay
    • LearnRateExpDecay
    • LearnRateInvTimeDecay
    • LearnRatePiecewiseConstDecay
    • LearnRatePolynomialDecay
• metrics
  • Accuracy
  • BaseMetric
  • BinaryCrossentropy
  • JSDivergence
  • KLDivergence
  • MeanAbsoluteError
  • MeanSquaredError
  • RootMeanSquaredError
  • WassertsteinDistance
• optimization
  • callbacks
    • HopaasPruner
  • scores
    • BaseScore
    • EMDistance
    • KSDistance
• players
  • classifiers
    • AuxClassifier
    • AuxMultiClassifier
    • Classifier
    • MultiClassifier
    • ResClassifier
    • ResMultiClassifier
  • discriminators
    • AuxDiscriminator
    • Discriminator
    • ResDiscriminator
  • generators
    • Generator
    • ResGenerator
• utils
  • checks
    • checkMetrics
    • checkOptimizer
  • preprocessing
    • invertColumnTransformer - 🐛
  • reports
    • HPSingleton
    • getSummaryHTML

🐛 Bug fixes

• invertColumnTransformer

  Problem. When the column indices passed to a transformer of the scikit-learn's ColumnTransformer aren't adjacent, this custom function has an unexpected behavior mixing the output columns.
  Solution. The function has been rewritten from scratch trying to follow a logical procedure that should mitigate new issues with the inversion of the ColumnTransformer.

🧩 Minor changes

Since regularization terms applied to either generator or discriminator can be extremely data-dependent, if they are computed also during the test step, it can produce loss values significantly different from the ones resulting in the train step. Hence, the GAN algorithms were updated so that the various _compute_*_loss methods take an additional boolean argument, called test, to avoid to compute any regularization terms during the test steps.

‼️ Note

This is the first release for Zenodo.