saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-05-22 22:01:08 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

[NNAPI] Minor fix (#15052)

Browse source 
### Description
<!-- Describe your changes. -->

Followed by https://github.com/microsoft/onnxruntime/pull/14881


### Motivation and Context
<!-- - Why is this change required? What problem does it solve?
- If it fixes an open issue, please link to the issue here. -->

---------

Co-authored-by: Edward Chen <18449977+edgchen1@users.noreply.github.com>
This commit is contained in:
JiCheng 2023-03-31 15:13:57 +08:00 committed by GitHub
parent d80859f63d
commit 60cc082f0a
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
2 changed files with 6 additions and 9 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

14
onnxruntime/core/providers/nnapi/nnapi_builtin/builders/model_builder.cc
View file

@ -543,7 +543,6 @@ Status ModelBuilder::Compile(std::unique_ptr<Model>& model) {
bool all_ops_supported = std::all_of(supported_ops, supported_ops + num_nnapi_ops_,
[](bool is_supported) { return is_supported; });
// TODO, To allow fallback to CPU if it's not strict CPU_DISABLED mode
if (!all_ops_supported) {
// There are some ops not supported by the list of the target devices
// Fail the Compile
@ -551,11 +550,10 @@ Status ModelBuilder::Compile(std::unique_ptr<Model>& model) {
// TODO, add some logic to not fail for some cases
// Such as, if there are some acceptable fall back to CPU (nnapi-reference)
// and CPU is not in the target devices list
return ORT_MAKE_STATUS(ONNXRUNTIME, EP_FAIL,
"The model cannot run using the current set of target devices, ",
GetDevicesDescription(nnapi_target_devices_));
return ORT_MAKE_STATUS(ONNXRUNTIME, EP_FAIL,
"The model cannot run using the current set of target devices, ",
GetDevicesDescription(nnapi_target_devices_));
}
// Workaround bugs in NNAPI drives on some phones
// where ops are passed checking by 'ANeuralNetworksModel_getSupportedOperationsForDevices'
// but failed at compilation.
@ -587,12 +585,12 @@ Status ModelBuilder::Compile(std::unique_ptr<Model>& model) {
optype_support_status[stat_name].second++;
}
}
size_t total_ops = 0;
size_t fallback_ops = 0;
std::string fallback_op_detail, normal_op_detail;
for (const auto& [op, ops_status] : optype_support_status) {
auto& [support_cnt, unspport_cnt] = ops_status;
total_ops += support_cnt + unspport_cnt;
fallback_ops += unspport_cnt;
if (support_cnt > 0) {
normal_op_detail += MakeString(support_cnt, "x ", op, ", ");
}
@ -601,7 +599,7 @@ Status ModelBuilder::Compile(std::unique_ptr<Model>& model) {
}
}
LOGS_DEFAULT(VERBOSE) << total_ops << " Ops [" << fallback_op_detail << "] out of " << num_nnapi_ops_
LOGS_DEFAULT(VERBOSE) << fallback_ops << " Ops [" << fallback_op_detail << "] out of " << num_nnapi_ops_
<< " are falling-back to " << kNnapiCpuDeviceName << ", and ["
<< normal_op_detail << "] is running in accelerators.";
}

1
onnxruntime/core/providers/nnapi/nnapi_builtin/builders/model_builder.h
View file

@ -5,7 +5,6 @@
#include <onnx/onnx_pb.h>
#include <unordered_set>
#include "core/common/inlined_containers.h"
#include "core/common/inlined_containers_fwd.h"
#include "core/graph/basic_types.h"
#include "core/providers/nnapi/nnapi_builtin/model.h"