Softplus - 1 vs 22

Next section compares an older to a newer version of the same operator after both definition are converted into markdown text. Green means an addition to the newer version, red means a deletion. Anything else is unchanged.

Files changed (1) hide show

1. Softplus1 → Softplus22 +1 -1

Softplus1 → Softplus22 RENAMED Viewed

	@@ -1 +1 @@
1 1	Softplus takes one input data (Tensor<T>) and produces one output data
2 2	(Tensor<T>) where the softplus function, y = ln(exp(x) + 1), is applied to
3 3	the tensor elementwise.
4 4	#### Function Body
5 5	The function definition for this operator.
6 6	<
7 7	domain: "",
8 8	opset_import: ["" : 18]
9 9	>
10 10	Softplus (X) => (Y)
11 11	{
12 12	exp_x = Exp (X)
13 13	one = Constant <value: tensor = float {1}> ()
14 14	one_cast = CastLike (one, X)
15 15	exp_x_add_one = Add (exp_x, one_cast)
16 16	Y = Log (exp_x_add_one)
17 17	}
18 18	### Inputs
19 19	- **X** (heterogeneous) - **T**:
20 20	1D input tensor
21 21	### Outputs
22 22	- **Y** (heterogeneous) - **T**:
23 23	1D input tensor
24 24	### Type Constraints
25	- * **T** in ( tensor(double), tensor(float), tensor(float16) ):
25	+ * **T** in ( tensor(bfloat16), tensor(double), tensor(float), tensor(float16) ):
26 26	Constrain input and output types to float tensors.