potok/opencv
1
0
Fork 0
mirror of https://github.com/opencv/opencv.git synced 2026-07-31 08:13:04 +04:00
Code Packages Releases Wiki Activity

Further optimization of Conv2D, fused Conv_Add_Activation, bring latest code from ficus OpConv.fx. (#22401)

Browse Source
This commit is contained in:
Zihao Mu
2022-08-26 17:57:25 +08:00
committed by GitHub
parent 67fa8a2f47
commit bb64db98d8
12 changed files with 1194 additions and 905 deletions
Download Patch File Download Diff File Expand all files Collapse all files
modules/dnn/test/test_caffe_importer.cpp
+1 -1
View File
@@ -284,7 +284,7 @@ TEST(Reproducibility_SSD, Accuracy)
Mat out = net.forward("detection_out");
Mat ref = blobFromNPY(_tf("ssd_out.npy"));
normAssertDetections(ref, out, "", FLT_MIN);
normAssertDetections(ref, out, "", 0.06);
}
typedef testing::TestWithParam<tuple<Backend, Target> > Reproducibility_MobileNet_SSD;
modules/dnn/test/test_int8_layers.cpp
+1 -1
View File
@@ -1029,7 +1029,7 @@ TEST_P(Test_Int8_nets, FasterRCNN_resnet50)
Mat blob = blobFromImage(inp, 1.0, Size(800, 600), Scalar(), true, false);
Mat ref = blobFromNPY(_tf("tensorflow/faster_rcnn_resnet50_coco_2018_01_28.detection_out.npy"));
float confThreshold = 0.5, scoreDiff = 0.05, iouDiff = 0.15;
float confThreshold = 0.8, scoreDiff = 0.05, iouDiff = 0.15;
testDetectionNet(net, blob, ref, confThreshold, scoreDiff, iouDiff);
}
modules/dnn/test/test_torch_importer.cpp
+2 -2
View File
@@ -488,7 +488,7 @@ TEST_P(Test_Torch_nets, ENet_accuracy)
// Due to numerical instability in Pooling-Unpooling layers (indexes jittering)
// thresholds for ENet must be changed. Accuracy of results was checked on
// Cityscapes dataset and difference in mIOU with Torch is 10E-4%
normAssert(ref, out, "", 0.00044, /*target == DNN_TARGET_CPU ? 0.453 : */0.552);
normAssert(ref, out, "", 0.0005, /*target == DNN_TARGET_CPU ? 0.453 : */0.552);
normAssertSegmentation(ref, out);
const int N = 3;
@@ -496,7 +496,7 @@ TEST_P(Test_Torch_nets, ENet_accuracy)
{
net.setInput(inputBlob, "");
Mat out = net.forward();
normAssert(ref, out, "", 0.00044, /*target == DNN_TARGET_CPU ? 0.453 : */0.552);
normAssert(ref, out, "", 0.0005, /*target == DNN_TARGET_CPU ? 0.453 : */0.552);
normAssertSegmentation(ref, out);
}
}