those that have is_input = True).Ĭomputes the gradient of this Function at location at with respect to wrt. Provided in arguments that correspond to each input Variable of You expect multiple functions to be returned, useĬomputes the values of speficied variables in outputs, using values Throws an exception if name occurs multiple times. Returns a primitive function with name in the graph starting from If you expect only one function to be returned, use Returns a list of primitive function with name in the graph In as_block, using the supplied op_name and name parameters, which are otherwise ignored. If BlockFunction() passes True, then the result will be wrapped Make_block=True is an internal parameter used to implement BlockFunction. If you are working with Python 2.7, use CNTK's Signature decorator instead:ĮlementTimes(x: Tensor) -> Tensor If you use Python 3, Functions with types are declared using Python annotation syntax, f(x:Tensor): In this case, the decorator creates a CNTK Function To train a Function or pass data to it, you need to declare the types Such a function can only be combined with other symbolic functions. The above form creates a CNTK Function whose arguments are placeholder variables. The Function's input signature is defined by the lambda. If it has only one output, one can invoke Variable methods on it, which itįunction objects can also be constructed directly from a Python lambda, Base class of all primitive tensor operators.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |