saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-07-11 17:48:34 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions
onnxruntime/onnxruntime/core
History
gwang-msft ca0dd8246c
NNAPI EP, add quantization support (#4530) 
* nnapi quantization work, 1. add SymmPerChannelQuantParams to operand types

* add ways for operand_type to change dimension

* remove per-channel quantization

* Squashed commit of the following:

commit 4857c3a732298c2f96efb61800b7621251d61c1b
Author: gwang <wanggy@outlook.com>
Date:   Tue Jul 14 23:02:28 2020 -0700

    remove per-channel quantization

commit 775e4b2960f531496b8d11eef27d64e5b85c3c66
Author: gwang <wanggy@outlook.com>
Date:   Mon Jul 13 11:54:02 2020 -0700

    add ways for operand_type to change dimension

commit e56a494de67c66f8122d908270fbc2bb17e38423
Author: gwang <wanggy@outlook.com>
Date:   Wed Jul 8 15:18:55 2020 -0700

    nnapi quantization work, 1. add SymmPerChannelQuantParams to operand types

* add support for QuantizeLinear

* add dequantizelinear support

* minor style update

* minor bug fix

* add quantization support for qlinearmatmul, minor issue fix

* add quantized input support, minor bug fix

* fix issues in the qlinearmatmul

* add verify scale and zeropoint for qlinearmatmul

* add test for [de]qunatizelinear ops

* add qlinearconv support

* fixed small issue causing test failure

* fix test exception

* fix for centos test failure

* fix centos test failure

* fix issue causing win-tensorRT ci failure

* addressed comments
2020-07-20 16:43:31 -07:00
..
codegen
common adding concat logic when initial path is empty (#4525) 2020-07-16 23:46:12 -07:00
dll
framework add Output() overloads (#4546) 2020-07-19 15:21:12 -07:00
graph Silence binscope warnings (#4542) 2020-07-20 14:10:41 -07:00
language_interop_ops
mlas fix xcode alerts (#4470) 2020-07-16 10:20:34 -07:00
optimizer Handle nodes with dtype attribute causing diffs in inferred and actual types of NodeArgs due to InsertCastTransformer (#4523) 2020-07-16 21:59:14 -07:00
platform fix xcode alerts (#4470) 2020-07-16 10:20:34 -07:00
profile Use OrtMutex and OrtCondVar everywhere instead of std::mutex/std::condition_variable for consistency. 2020-06-03 08:42:16 -07:00
protobuf
providers NNAPI EP, add quantization support (#4530) 2020-07-20 16:43:31 -07:00
session Support training_mode flag in eval (#4324) 2020-07-08 10:38:54 -07:00
util MLAS: support prepacking APIs for quantized GEMM (#4433) 2020-07-06 15:20:10 -07:00