TF 2.0 is out! Get hands-on practice at TF World, Oct 28-31. Use code TF20 for 20% off select passes. Register now

Module: tfp.experimental.substrates.numpy.math.numeric

View source on GitHub

Numerically stable variants of common mathematical expressions.

Functions

clip_by_value_preserve_gradient(...): Clips values to a specified min and max while leaving gradient unaltered.

log1psquare(...): Numerically stable calculation of log(1 + x**2) for small or large |x|.