to `tf.setdiff1d` to match NumPy naming. * `tf.inv` has been renamed to be
`tf.reciprocal` (component-wise reciprocal) to avoid confusion with `np.inv`
which is matrix inversion * tf.round now uses banker's rounding (round to even)
semantics to match NumPy. * `tf.split` now takes arguments in a reversed order
and with different keywords. In particular, we now match NumPy order as

x.reset(len(y.limbs)) copy(x.limbs, y.limbs) return x } // Bits returns x as a little-endian slice of uint. The length of the slice // matches the announced length of x. The result and x share the same underlying // array. func (x *Nat) Bits() []uint { return x.limbs } // Bytes returns x as a zero-extended big-endian byte slice. The size of the // slice will match the size of m. // // x must have the same size as m and it must be less than or equal to m. func (x *Nat) Bytes(m *Modulus) []byte { i...

x.reset(len(y.limbs)) copy(x.limbs, y.limbs) return x } // Bits returns x as a little-endian slice of uint. The length of the slice // matches the announced length of x. The result and x share the same underlying // array. func (x *Nat) Bits() []uint { return x.limbs } // Bytes returns x as a zero-extended big-endian byte slice. The size of the // slice will match the size of m. // // x must have the same size as m and it must be less than or equal to m. func (x *Nat) Bytes(m *Modulus) []byte { i...