Add type info to MatMul spec (#4855)

This commit is contained in:
Katarzyna Mitrus 2021-03-22 08:28:58 +01:00 committed by GitHub
parent a868d88d49
commit 3866896ba6
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23

View File

@ -57,14 +57,17 @@ Two attributes, `transpose_a` and `transpose_b` specify embedded transposition f
**Inputs**:
* **1**: Input batch of matrices A. Rank >= 1. Required.
* **1**: Tensor of type *T* with matrices A. Rank >= 1. Required.
* **2**: Input batch of matrices B. Rank >= 1. Required.
* **2**: Tensor of type *T* with matrices B. Rank >= 1. Required.
**Outputs**
* **1**: Tensor with results of the multiplication.
* **1**: Tensor of type *T* with results of the multiplication.
**Types**:
* *T*: any supported floating point or integer type.
**Example**