commit | 044225edc558796a86e0e646e3fed3696c8c34c8 | [log] [tgz] |
---|---|---|
author | Miao Wang <miaowang@google.com> | Wed Jul 15 11:38:04 2015 -0700 |
committer | Miao Wang <miaowang@google.com> | Thu Jul 16 10:27:51 2015 -0700 |
tree | d56393446585a1c20dd219dddab9734d5a1cf863 | |
parent | 124d389d0192834ab1cb7e020b9c37cc2b36d01a [diff] |
[RenderScript] fix BNNM docs for support lib bug: 22507076 Change-Id: I864e2c7e32f55c501db80d1f28081800fab15027 (cherry picked from commit 513bce7b3def28727698a86e8ded969f1be3351a)
diff --git a/v8/renderscript/java/src/android/support/v8/renderscript/ScriptIntrinsicBLAS.java b/v8/renderscript/java/src/android/support/v8/renderscript/ScriptIntrinsicBLAS.java index 1997c80..9f936b1 100644 --- a/v8/renderscript/java/src/android/support/v8/renderscript/ScriptIntrinsicBLAS.java +++ b/v8/renderscript/java/src/android/support/v8/renderscript/ScriptIntrinsicBLAS.java
@@ -4134,7 +4134,7 @@ /** - * 8-bit GEMM-like operation for neural networks: C = B.transposed() * A + * 8-bit GEMM-like operation for neural networks: C = A * Transpose(B) * Calculations are done in 1.10.21 fixed-point format for the final output, * just before there's a shift down to drop the fractional parts. The output * values are gated to 0 to 255 to fit in a byte, but the 10-bit format