covdx: Covariance between derivatives and hidden states
In mgoulet847/tagi: Tractable Approximate Gaussian Inference in Neural Networks

This function calculates covariance between derivatives and hidden states. It is not related to the derivative calculation process. It could be used infer Z (hidden states) with the constraint that the derivative of g with respect to Z equals 0.

1	covdx(mwo, mw, mdgo2, mpdi, mdgoe, Cdozi, Cdizi, ni, no, no2, B)

`mwo`	Mean vector of weights for the next layer
`mw`	Mean vector of weights for the current layer
`mdgo2`	Mean vector of product of derivatives in second next layer
`mpdi`	Mean vector of first derivative product wd of current layer
`mdgoe`	Mean of product of derivatives at each node in next layer
`Cdozi`	Covariance between derivative (next) and hidden (current) layers
`Cdizi`	Covariance between derivative and hidden layers (same layer)
`ni`	Number of units in current layer
`no`	Number of units in next layer
`no2`	Number of units in second next layer
`B`	Batch size