Pr3D - PreDicting Delay probability Density

Implementation of two conditional density estimation methods with parametric neural networks in Python using Tensorflow

• Conventional and non-conditional mixture density network with Gaussian mixture model (GaussianMM)
• Novel conventional and non-conditional mixture density network with Gamma and extreme value mixture model (GammaEVM)

Using the package

pip install git+https://github.com/samiemostafavi/pr3d.git

Helpful readings

[1] https://keras.io/examples/keras_recipes/bayesian_neural_networks/

[2] https://towardsdatascience.com/modeling-uncertainty-in-neural-networks-with-tensorflow-probability-a706c2274d12

[3] https://nnart.org/understanding-a-bayesian-neural-network-a-tutorial/

[4] https://towardsdatascience.com/bayesian-neural-networks-with-tensorflow-probability-fbce27d6ef6

[5] https://towardsdatascience.com/data-formats-for-training-in-tensorflow-parquet-petastorm-feather-and-more-e55179eeeb72

[6] https://www.tensorflow.org/probability/api_docs/python/tfp/layers/DenseVariational

[7] https://stackoverflow.com/questions/58678836/notimplementederror-layers-with-arguments-in-init-must-override-get-conf

[8] https://www.youtube.com/watch?v=VFEOskzhhbc