ANSCENTER/ANSCORE
2
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Fix AMD by changing from GetTensorData<T>() to GetTensorMutableData<T>()

Browse Source
This commit is contained in:
Tuan Nghia Nguyen
2026-04-28 13:25:02 +10:00
parent f4b74c837e
commit dcf974c35c
18 changed files with 359 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
modules/ANSODEngine/ANSYOLO12OD.cpp
View File

@@ -518,14 +518,16 @@ namespace ANSCENTER {
}
}
std::vector<Object> YOLO12OD::postprocess(const cv::Size& originalImageSize, const cv::Size& resizedImageShape,
const std::vector<Ort::Value>& outputTensors,
float confThreshold, float iouThreshold)
std::vector<Ort::Value>& outputTensors,
float confThreshold, float iouThreshold)
{
std::lock_guard<std::recursive_mutex> lock(_mutex);
try {
std::vector<Object> detections;
const float* rawOutput = outputTensors[0].GetTensorData<float>(); // Extract raw output data from the first output tensor
// GetTensorMutableData (not GetTensorData) on DML — const variant
// hangs on AMD after ~8 calls. Read-only despite the name.
const float* rawOutput = outputTensors[0].GetTensorMutableData<float>();
const std::vector<int64_t> outputShape = outputTensors[0].GetTensorTypeAndShapeInfo().GetShape();
// Determine the number of features and detections