1
0
mirror of https://github.com/opencv/opencv.git synced 2026-07-30 07:43:03 +04:00

Merge pull request #24120 from dkurt:actualize_dnn_links

OCL_FP16 MatMul with large batch

* Workaround FP16 MatMul with large batch

* Fix OCL reinitialization

* Higher thresholds for INT8 quantization

* Try fix gemm_buffer_NT for half (columns)

* Fix GEMM by rows

* Add batch dimension to InnerProduct layer test

* Fix Test_ONNX_conformance.Layer_Test/test_basic_conv_with_padding

* Batch 16

* Replace all vload4

* Version suffix for MobileNetSSD_deploy Caffe model
This commit is contained in:
Dmitry Kurtaev
2023-08-16 15:46:11 +03:00
committed by GitHub
parent 8d1c73a912
commit 8ad5eb521a
10 changed files with 49 additions and 47 deletions
+2 -2
View File
@@ -101,8 +101,8 @@ PERF_TEST(SqueezeNet_v1_1_caffe, CaffePerfTest)
PERF_TEST(MobileNet_SSD, CaffePerfTest)
{
caffe::Net<float>* net = initNet("dnn/MobileNetSSD_deploy.prototxt",
"dnn/MobileNetSSD_deploy.caffemodel");
caffe::Net<float>* net = initNet("dnn/MobileNetSSD_deploy_19e3ec3.prototxt",
"dnn/MobileNetSSD_deploy_19e3ec3.caffemodel");
TEST_CYCLE() net->Forward();
SANITY_CHECK_NOTHING();
}