Evolution-x/hardware_interfaces
1
0
Fork 0
mirror of https://github.com/Evolution-X/hardware_interfaces synced 2026-02-02 06:22:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fix the spec for TENSOR_QUANT8_ASYMM to match our validation.

Browse Source 
- Scale of 0.0 is invalid for quantized tensor.

Bug: 77236592
Test: mma
Merged-In: I3a53d6303d8c964d451e17a3b1671de82d0ff335
Change-Id: I3a53d6303d8c964d451e17a3b1671de82d0ff335
(cherry picked from commit a82d39102a)
This commit is contained in:
Miao Wang
2018-04-05 16:27:00 -07:00
committed by Michael Butler
parent 9237ae8889
commit e12088cd64
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
current.txt
View File

@@ -248,5 +248,5 @@ c8bc853546dd55584611def2a9fa1d99f657e3366c976d2f60fe6b8aa6d2cb87 android.hardwar
# Future changes to HALs
5804ca86611d72e5481f022b3a0c1b334217f2e4988dad25730c42af2d1f4d1c android.hardware.neuralnetworks@1.0::IDevice
08ae9fc24f21f809e9b8501dfbc803662fcd6a8d8e1fb71d9dd7c0c4c6f5d556 android.hardware.neuralnetworks@1.0::types
1488db5ffb8a7979488d1084761aab8bca2f59bc9a02d75cdefc296afeaf591b android.hardware.neuralnetworks@1.0::types

2
neuralnetworks/1.0/types.hal
View File

@@ -44,7 +44,7 @@ enum OperandType : int32_t {
*
* Attached to this tensor are two numbers that can be used to convert the
* 8 bit integer to the real value and vice versa. These two numbers are:
* - scale: a 32 bit floating point value
* - scale: a 32 bit floating point value greater than zero
* - zero_value: a 32 bit integer
*
* The formula is: