Where Is `*` Documented In Tensorflow?

I don't find where * is documented. It seems that it can be equivalent to either tf.multiply or tf.scalar_mul. Is it so?

Solution 1:

The most reliable documentation is the source code:

def _mul_dispatch(x, y, name=None):
  """Dispatches cwise mul for "Dense*Dense" and "Dense*Sparse"."""
  is_tensor_y = isinstance(y, ops.Tensor)
  if is_tensor_y:
    return gen_math_ops._mul(x, y, name=name)
  else:
    assert isinstance(y, sparse_tensor.SparseTensor)  # Case: Dense * Sparse.
    new_vals = gen_sparse_ops.sparse_dense_cwise_mul(y.indices, y.values,
                                                     y.dense_shape, x, name)
    return sparse_tensor.SparseTensor(y.indices, new_vals, y.dense_shape)

...

_OverrideBinaryOperatorHelper(_mul_dispatch, "mul")

This means __mul__ operator overload, which does _mul_dispatch. As you can see, it calls either gen_math_ops._mul (which is the under-the-hood core function of tf.multiply) or sparse_dense_cwise_mul if the tensor is sparse.

By the way, tf.scalar_mul is just a wrapper over scalar * x (source code), so it's basically the same thing, but the dependency is the other way around.