f3c511060ce62072a34a50319237d28b82d10448,gpflow/base.py,Parameter,__init__,#,88

Before Change

        self.prior = prior
        self.prior_on = prior_on  // type: ignore  // see https://github.com/python/mypy/issues/3004

        if isinstance(value, tf.Variable):
            self._unconstrained = value
        else:
            unconstrained_value = self.validate_unconstrained_value(value, dtype)
            self._unconstrained = tf.Variable(
                unconstrained_value, dtype=dtype, name=name, trainable=trainable
            )

    def log_prior_density(self) -> tf.Tensor:
         Log of the prior probability density of the constrained variable. 

        if self.prior is None:

After Change

        therefore we need a positive constraint and it is natural to use constrained values.
        A prior can be imposed either on the constrained version (default) or on the unconstrained version of the parameter.
        
        if transform is None:
            transform = tfp.bijectors.Identity()

        value = _cast_to_dtype(value, dtype)
        _validate_unconstrained_value(value, transform, dtype)
        super().__init__(value, transform, dtype=value.dtype, trainable=trainable, name=name)

Instances