gfdl.weights.lecun_normal

lecun_normal(d, h, *, rng, **kwargs)

The weight function returning samples drawn from Lecun normal distribution.

Parameters:

dint

Number of features in a sample or number of neurons in the previous hidden layer.

hint

Number of neurons in the current hidden layer.

rngnp.random.Generator

A NumPy random number generator instance.

Returns:

ndarray or scalar

Draw samples from the Lecun normal distribution with mean 0 and standard deviation sqrt(1/h).

Other Parameters:

**kwargsdict

Needed for keyword arguments and compatibility with other weight function apis but not relevant for this function.

Notes

https://www.tensorflow.org/api_docs/python/tf/keras/initializers/LecunNormal