A machine learning model that uses probabilistic layers regression, modelled using TensorFlow. Probabilistic layers regression is a type of linear regression in which the neurons of the neural network are actually probability calculations. We use the TensorFlow Probability module, and consider 5 cases:
- There is no uncertainty in the data
- There is aleatoric uncertainty in the data
- There is epistemic uncertainty in the data
- There is both aleatoric and epistemic uncertainty in the data
- There is functional uncertainty in the data NumPy has been used in the storage of data and Matplotlib has been used to plot the figures