Comments (5)
qianxiaoer
commented on July 17, 2024
哈罗，邮件收到了，我会尽快查看，回复，谢谢！
from hyperlpr.
zxm97
commented on July 17, 2024
编译了MNN2.7.1 (设置MNN_CUDA=ON) 后，编译本项目，CMakeLists的内容修改之处：
option( LINUX_FETCH_MNN "Fetch and build MNN from git" OFF )
option( LINUX_USE_3RDPARTY_OPENCV "Linux platform using pre-compiled OpenCV library from 3rdparty_hyper_inspire_op" OFF)
option( BUILD_SHARE "Build shared libs" ON )
option( BUILD_SAMPLES "Build samples demo" OFF )
option( BUILD_TEST "Build unit-test exec" OFF )
...
if (LINUX_FETCH_MNN)
...
else()
# MNN Third party dependence
set(MNN_INCLUDE_DIRS ${PATH_3RDPARTY}/MNN-2.7.1/${PLAT}/include)
set(MNN_LIBS ${PATH_3RDPARTY}/MNN-2.7.1/${PLAT}/lib)
endif()
HyperLPR-master\cpp\src\nn_module\mnn_adapter.h (21)的is_use_cuda设为false或true编译完都检测不到。
用下载的第三方库里的MNN-2.2.0，也没检测到。
from hyperlpr.
zxm97
commented on July 17, 2024
nano@nano-desktop:~/HyperLPR-master/build_test$ ./UnitTest
[test_classification.cpp][C_A_T_C_H_T_E_S_T_0][17]: [UnitTest]->Classification Model
The device support i8sdot:0, support fp16:0, support i8mm: 0
UnitTest is a Catch v2.13.9 host application.
Run with -? for options
-------------------------------------------------------------------------------
test_Classification
test_ClassificationModelPredict
-------------------------------------------------------------------------------
/home/nano/HyperLPR-master/cpp/test/nn_module/test_classification.cpp:40
...............................................................................
/home/nano/HyperLPR-master/cpp/test/nn_module/test_classification.cpp:49: FAILED:
CHECK( clsEngine.getMOutputMaxConfidence() == Approx(predict_results_confidence[i]).epsilon(0.001) )
with expansion:
0.87586f == Approx( 0.9999293089 )
/home/nano/HyperLPR-master/cpp/test/nn_module/test_classification.cpp:48: FAILED:
CHECK( PlateColor(clsEngine.getMOutputColor()) == predict_results_cls[i] )
with expansion:
0 == 2
/home/nano/HyperLPR-master/cpp/test/nn_module/test_classification.cpp:49: FAILED:
CHECK( clsEngine.getMOutputMaxConfidence() == Approx(predict_results_confidence[i]).epsilon(0.001) )
with expansion:
0.99349f == Approx( 0.8975974917 )
/home/nano/HyperLPR-master/cpp/test/nn_module/test_classification.cpp:48: FAILED:
CHECK( PlateColor(clsEngine.getMOutputColor()) == predict_results_cls[i] )
with expansion:
0 == 1
/home/nano/HyperLPR-master/cpp/test/nn_module/test_classification.cpp:49: FAILED:
CHECK( clsEngine.getMOutputMaxConfidence() == Approx(predict_results_confidence[i]).epsilon(0.001) )
with expansion:
0.9302f == Approx( 0.9997951984 )
===============================================================================
[test_detection.cpp][C_A_T_C_H_T_E_S_T_0][15]: [UnitTest]->Detect Model
[test_detection.cpp][C_A_T_C_H_T_E_S_T_0][25]: Detect Model SplitModel
-------------------------------------------------------------------------------
test_Detection
test_SplitDetectionSplitModel
-------------------------------------------------------------------------------
/home/nano/HyperLPR-master/cpp/test/nn_module/test_detection.cpp:24
...............................................................................
/home/nano/HyperLPR-master/cpp/test/nn_module/test_detection.cpp:32: FAILED:
CHECK( result.size() == 1 )
with expansion:
0 == 1
/home/nano/HyperLPR-master/cpp/test/nn_module/test_detection.cpp:32: FAILED:
{Unknown expression after the reported line}
due to a fatal error condition:
SIGSEGV - Segmentation violation signal
===============================================================================
test cases: 2 | 0 passed | 2 failed
assertions: 26 | 19 passed | 7 failed
Segmentation fault (core dumped)
from hyperlpr.
zxm97
commented on July 17, 2024
MNN 2.7.1单元测试的结果：
nano@nano-desktop:~/MNN-2.7.1/build$ ./run_test.out
running core/auto_storage.
running engine/backend/copy_buffer_float.
The device support i8sdot:0, support fp16:0, support i8mm: 0
Test 0 Backend for 0
res=1 in run, 672
========= check nchwTonhwc result ! =========
res=1 in run, 674
========= check nhwc_2_NC4HW4_2_nhwc_float result ! =========
NC4HW4 -> nhwc !
res=1 in run, 676
res=1 in run, 678
res=1 in run, 680
res=1 in run, 682
Test 0 Backend for 1
res=1 in run, 672
========= check nchwTonhwc result ! =========
res=1 in run, 674
========= check nhwc_2_NC4HW4_2_nhwc_float result ! =========
NC4HW4 -> nhwc !
res=1 in run, 676
res=1 in run, 678
res=1 in run, 680
res=1 in run, 682
Test 0 Backend for 2
res=1 in run, 672
========= check nchwTonhwc result ! =========
res=1 in run, 674
========= check nhwc_2_NC4HW4_2_nhwc_float result ! =========
NC4HW4 -> nhwc !
res=1 in run, 676
res=1 in run, 678
res=1 in run, 680
res=1 in run, 682
Test 2 Backend for 0
res=1 in run, 672
========= check nchwTonhwc result ! =========
res=1 in run, 674
========= check nhwc_2_NC4HW4_2_nhwc_float result ! =========
NC4HW4 -> nhwc !
res=1 in run, 676
res=1 in run, 678
res=1 in run, 680
res=1 in run, 682
Test 2 Backend for 1
res=1 in run, 672
========= check nchwTonhwc result ! =========
res=1 in run, 674
========= check nhwc_2_NC4HW4_2_nhwc_float result ! =========
NC4HW4 -> nhwc !
res=1 in run, 676
res=1 in run, 678
res=1 in run, 680
res=1 in run, 682
Test 2 Backend for 2
res=1 in run, 672
========= check nchwTonhwc result ! =========
res=1 in run, 674
========= check nhwc_2_NC4HW4_2_nhwc_float result ! =========
NC4HW4 -> nhwc !
res=1 in run, 676
res=1 in run, 678
res=1 in run, 680
res=1 in run, 682
running engine/backend/copy_buffer_cpu.
Test 0 Backend for 0
========= check NC4HW4_2_NC4HW4_IntType result ! =========
========= check NC4HW4_2_NC4HW4_IntType result ! =========
========= check NC4HW4_2_NC4HW4_IntType result ! =========
========= check nhwc_2_NC4HW4_2_nhwc_inttype result ! =========
========= check nhwc_2_NC4HW4_2_nhwc_inttype result ! =========
========= check nhwc_2_NC4HW4_2_nhwc_inttype result ! =========
Test 0 Backend for 1
========= check NC4HW4_2_NC4HW4_IntType result ! =========
========= check NC4HW4_2_NC4HW4_IntType result ! =========
========= check NC4HW4_2_NC4HW4_IntType result ! =========
========= check nhwc_2_NC4HW4_2_nhwc_inttype result ! =========
========= check nhwc_2_NC4HW4_2_nhwc_inttype result ! =========
========= check nhwc_2_NC4HW4_2_nhwc_inttype result ! =========
Test 0 Backend for 2
========= check NC4HW4_2_NC4HW4_IntType result ! =========
========= check NC4HW4_2_NC4HW4_IntType result ! =========
========= check NC4HW4_2_NC4HW4_IntType result ! =========
========= check nhwc_2_NC4HW4_2_nhwc_inttype result ! =========
========= check nhwc_2_NC4HW4_2_nhwc_inttype result ! =========
========= check nhwc_2_NC4HW4_2_nhwc_inttype result ! =========
running core/buffer_allocator.
BufferAllocator total size : 27 B, 0.000026 M
StaticAllocator total size : 0 B, 0.000000 M
BufferAllocator total size : 1189085472 B, 1134.000244 M
StaticAllocator total size : 0 B, 0.000000 M
BufferAllocator total size : 2776896 B, 2.648254 M
StaticAllocator total size : 0 B, 0.000000 M
running core/callback.
running core/idst.
running core/memory_utils.
running core/regionfuse.
running core/tensor.
running core/tensor_utils.
running core/threadpool.
workIndex=0 in operator(), 26
workIndex=1 in operator(), 26
workIndex=-1 in operator(), 26
index=0 in operator(), 29
index=0 in operator(), 29
index=0 in operator(), 29
index=4 in operator(), 29
index=4 in operator(), 29
index=8 in operator(), 29
index=8 in operator(), 29
workIndex=-1 in operator(), 26
index=0 in operator(), 29
index=1 in operator(), 29
index=2 in operator(), 29
index=3 in operator(), 29
index=4 in operator(), 29
index=5 in operator(), 29
index=6 in operator(), 29
index=7 in operator(), 29
index=8 in operator(), 29
index=9 in operator(), 29
index=1 in operator(), 29
workIndex=-1 in operator(), 26
index=0 in operator(), 29
index=2 in operator(), 29
index=1 in operator(), 29
index=3 in operator(), 29
index=2 in operator(), 29
index=4 in operator(), 29
index=3 in operator(), 29
index=5 in operator(), 29
index=4 in operator(), 29
index=6 in operator(), 29
index=5 in operator(), 29
index=7 in operator(), 29
index=1 in operator(), 29
index=2 in operator(), 29
index=6 in operator(), 29
index=2 in operator(), 29
index=6 in operator(), 29
index=5 in operator(), 29
index=9 in operator(), 29
index=1 in operator(), 29
index=5 in operator(), 29
index=9 in operator(), 29
workIndex=-1 in operator(), 26
index=0 in operator(), 29
workIndex=-1 in operator(), 26
index=0 in operator(), 29
workIndex=-1 in operator(), 26
index=0 in operator(), 29
index=1 in operator(), 29
index=2 in operator(), 29
index=1 in operator(), 29
index=1 in operator(), 29
index=3 in operator(), 29
index=2 in operator(), 29
index=2 in operator(), 29
index=4 in operator(), 29
index=3 in operator(), 29
index=3 in operator(), 29
index=5 in operator(), 29
index=4 in operator(), 29
index=4 in operator(), 29
index=6 in operator(), 29
index=5 in operator(), 29
index=5 in operator(), 29
index=7 in operator(), 29
index=6 in operator(), 29
index=8 in operator(), 29
index=6 in operator(), 29
index=7 in operator(), 29
index=9 in operator(), 29
index=7 in operator(), 29
index=8 in operator(), 29
index=8 in operator(), 29
index=9 in operator(), 29
index=9 in operator(), 29
index=6 in operator(), 29
workIndex=-1 in operator(), 26
workIndex=-1 in operator(), 26
index=0 in operator(), 29
index=8 in operator(), 29
index=3 in operator(), 29
index=7 in operator(), 29
index=3 in operator(), 29
index=7 in operator(), 29
index=0 in operator(), 29
index=9 in operator(), 29
index=1 in operator(), 29
index=1 in operator(), 29
index=2 in operator(), 29
index=3 in operator(), 29
index=4 in operator(), 29
index=5 in operator(), 29
index=6 in operator(), 29
index=7 in operator(), 29
index=8 in operator(), 29
index=2 in operator(), 29
index=3 in operator(), 29
index=4 in operator(), 29
index=5 in operator(), 29
index=6 in operator(), 29
index=7 in operator(), 29
index=8 in operator(), 29
index=9 in operator(), 29
index=9 in operator(), 29
index=7 in operator(), 29
index=8 in operator(), 29
index=9 in operator(), 29
running cv/image_process/gray_to_gray.
running cv/image_process/gray_to_gray_bilinear_transorm.
running cv/image_process/gray_to_gray_nearest_transorm.
running cv/image_process/gray_to_rgba.
running cv/image_process/bgr_to_gray.
running cv/image_process/rgb_to_bgr.
running cv/image_process/rgba_to_bgra.
running cv/image_process/bgr_to_bgr.
running cv/image_process/rgb_to_gray.
running cv/image_process/rgba_to_gray.
running cv/image_process/rgba_to_gray_bilinear_transorm.
running cv/image_process/rgba_to_gray_nearest_transorm.
running cv/image_process/rgba_to_bgr.
running cv/image_process/bgr_to_bgr_blitter.
running cv/image_process/gray_to_gray_blitter.
running cv/image_process/yuv_blitter.
running cv/matrix/scale.
running expr/AllAny.
running expr/ExecutorReset.
running expr/ExecutorConfigTest.
running expr/ExecutorScopeMultiThread.
running expr/ExprResizeCompute.
running expr/ExprResize.
running expr/Extra.
running expr/Gather.
running expr/MatMul.
running expr/MatrixBand.
running expr/MemoryIncrease/mobilenetv1.
From init 20.778854 mb to 20.778854 mb
running expr/MemoryIncrease/interp.
From init 1.469353 mb to 1.469353 mb
running expr/MidOutputTest.
running expr/ModuleTest.
Increase: 33.429974 in rt
Increase: 0.000000 in rt
running expr/RefTest.
running expr/LoopTest.
running expr/ModuleClone.
running expr/ModuleReleaseTest.
memory=f 0.000000 in run, 424
memory=f 66.859947 in run, 437
memory=f 0.000000 in run, 440
running expr/ModuleTestSpeed.
Thread 1 avg cost: 195.231812 ms
Thread 4 avg cost: 136.004807 ms
running expr/SpecialSessionTest.
running expr/SessionCircleTest.
loop: 1, 0.500000 -> 19.221680, 5.000000 -> 1922.167969
loop: 0, 0.500000 -> 0.500000, 5.000000 -> 50.000000
running expr/SessionTest.
_run, 705, cost time: 907.140015 ms
_run, 714, cost time: 710.296021 ms
operator(), 731, cost time: 1659.537109 ms
operator(), 731, cost time: 1978.366089 ms
operator(), 731, cost time: 2011.487061 ms
operator(), 731, cost time: 1856.212036 ms
_run, 758, cost time: 851.844055 ms
_run, 758, cost time: 644.807007 ms
_run, 758, cost time: 755.100037 ms
_run, 705, cost time: 911.361023 ms
_run, 714, cost time: 693.511047 ms
operator(), 731, cost time: 1612.624023 ms
operator(), 731, cost time: 1722.580078 ms
operator(), 731, cost time: 1733.859131 ms
operator(), 731, cost time: 1401.192017 ms
_run, 758, cost time: 788.224060 ms
_run, 758, cost time: 893.152039 ms
_run, 758, cost time: 708.292053 ms
running expr/MultiThreadOneSessionTest.
running expr/MemeoryUsageTest.
memory=f 56.751022 in operator(), 851
memory=f 24.000000 in operator(), 851
memory=f 7.667923 in operator(), 851
memory=f 11.444092 in operator(), 851
memory=f 15.258789 in operator(), 851
running expr/ConstMemoryReplaceTest.
running expr/MutlThreadConstReplaceTest.
Summer: 9.901236, 9.901236, 9.901236, 9.901236,
running expr/MultiThreadLoad.
running expr/Padding.
running expr/RasterOutput.
running expr/Replace.
running expr/Precompute.
running expr/PrecomputeDynamic.
running expr/ReverseSequence.
running expr/zeroshape.
running expr/zeroshape2.
running expr/zeroshape3.
running expr/zeroshape4.
running op/argmax.
running op/argmin.
running op/BatchMatMul.
running op/batch_to_space_nd.
running op/binary/broadcastShapeTest.
running op/binary/add.
running op/binary/subtract.
running op/binary/multiply.
running op/binary/divide.
running op/binary/pow.
running op/binary/minimum.
running op/binary/maximum.
running op/binary/biasadd.
running op/binary/greater.
running op/binary/greaterequal.
running op/binary/less.
running op/binary/floordiv.
running op/binary/squareddifference.
running op/binary/equal.
running op/binary/lessequal.
running op/binary/floormod.
running op/binary/mod_float.
running op/binary/mod_int.
running op/binary/atan2.
running op/binary/logicalor.
running op/binary/notqual.
running op/binary/subtractBroastTest.
running op/binary/bitwise_and.
running op/binary/bitwise_or.
running op/binary/bitwise_xor.
running op/binary/fuse_relu.
running op/binary/addInt8.
AddInt8 test zeropoint is zero
AddInt8 test zeropoint is not zero
running op/binary/subtractInt8.
SubtractInt8 test zeropoint is zero
SubtractInt8 test zeropoint is not zero
running op/binary/multiplyInt8.
MultiplyInt8 test zeropoint is zero
MultiplyInt8 test zeropoint is not zero
running op/binary/divideInt8.
DivedeInt8 test zero point is zero
DivedeInt8 test zero point is not zero
running op/binary/powInt8.
running op/binary/minimumInt8.
running op/binary/maximumInt8.
MaximumInt8 test zeropoint is zero
MaximumInt8 test zeropoint is not zero
running op/binary/floordivInt8.
running op/binary/floormodInt8.
running op/binary/atan2Int8.
running op/binary/sqdInt8.
SqdInt8 test zeropoint is zero
SqdInt8 test zeropoint is not zero
running op/binary/addC4.
running op/BroadcastToTest.
running op/BinaryBroadcastTest.
running op/cast.
running op/channel_shuffle.
running op/concat.
running op/Conv2DBackPropFilter.
running op/Conv2DBackPropFilterDW.
running op/Conv2DBackPropTest.
running op/bias_grad.
running op/ConvInt8/im2col_gemm.
running op/ConvInt8/im2col_spmm.
running op/ConvInt8/winograd.
running op/ConvInt8/depthwise.
Test strides=1
strides=2
running op/convert.
running op/convolution/conv3d.
running op/convolution/conv2d.
running op/convolution/sparse_conv2d.
running op/convolution/depthwise_conv.
running op/convolution/conv_group.
running op/CosineSimilarity.
running op/CropAndResize.
running op/crop.
running op/cumprod.
running op/cumsum.
running op/Deconvolution.
beigin testcase 0
beigin testcase 1
beigin testcase 2
running op/DeconvolutionInt8.
begin testcase 0
begin testcase 1
begin testcase 2
running op/depthtospace.
running op/Dilation2D/cpu.
running op/elu.
running op/expand_dims.
running op/fill.
running op/GatherElements.
running op/gather_nd.
running op/gather.
running op/gatherv2.
running op/GridSample.
running op/histogram.
running op/im2col.
running op/col2im.
running op/InnerProduct.
running op/layernorm.
running op/linspace.
running op/matmul.
running op/matmulBConst.
running op/matrixbandpart.
running op/moments.
running op/MultiConv.
running op/MultiDeconv.
running op/normalize.
running op/OneHotTest.
running op/prelu.
running op/pad.
running op/PermuteTest.
running op/MaxPool3d.
running op/AvePool3d.
running op/PoolGrad.
running op/ROIAlign.
running op/ROIPooling.
running op/randomuniform.
running op/range.
running op/rank.
running op/raster.
running op/relu6.
running op/clamp.
running op/relu.
running op/reduction/reduce_sum.
running op/reduction/reduce_sum_multi.
running op/reduction/reduce_mean.
running op/reduction/reduce_max.
running op/reduction/reduce_min.
running op/reduction/reduce_prod.
running op/reshape/nchw.
running op/reshape/nhwc.
running op/reshape/nc4hw4.
running op/resize.
running op/Interp.
running op/InterpInt8.
InterpInt8 test: Type=1
InterpInt8 test: Type=2
0 error, right: -1, compute: -0.032
InterpInt8 ResizeType=2 test failed!
running op/reverse.
running op/scale.
running op/scaleInt8.
running op/ScatterElementsTest.
running op/ScatterNdTest.
running op/selu.
running op/select.
running op/rnn/SequenceGRU.
running op/setdiff1d.
running op/shape.
running op/size.
running op/softmax.
running op/softmaxInt8.
running op/softplus.
running op/softsign.
running op/sort.
running op/space_to_batch_nd.
running op/spacetodepth.
running op/split.
running op/squeeze.
Cannot Squeeze dim[1], 1 is expected, 2 is got. input shape: Tensor shape: 2, 2,
running op/stack.
running op/stridedslice.
running op/splitc4.
running op/svd.
run, 37, cost time: 0.068000 ms
run, 136, cost time: 0.080000 ms
running op/tanh.
running op/threshold.
running op/tile.
running op/TopKV2.
0 s 74 ms
running op/transpose.
running op/unary/abs.
running op/unary/negative.
running op/unary/floor.
running op/unary/ceil.
running op/unary/square.
running op/unary/sqrt.
running op/unary/rsqrt.
running op/unary/exp.
running op/unary/log.
running op/unary/sin.
running op/unary/cos.
running op/unary/tan.
running op/unary/asin.
running op/unary/acos.
running op/unary/atan.
running op/unary/reciprocal.
running op/unary/log1p.
running op/unary/tanh.
running op/unary/sigmoid.
running op/unary/acosh.
running op/unary/asinh.
running op/unary/atanh.
running op/unary/round.
running op/unary/sign.
running op/unary/cosh.
running op/unary/erf.
running op/unary/erfc.
running op/unary/erfinv.
running op/unary/expm1.
running op/unary/sinh.
running op/unary/gelu.
running op/unique.
running op/UnravelIndexTest.
running op/unstack.
running op/where.
running op/zeroslike.
Error: op/InterpInt8
TEST_NAME_UNIT: 鍗曞厓娴嬭瘯
TEST_CASE_AMOUNT_UNIT: {"blocked":0,"failed":1,"passed":289,"skipped":0}
from hyperlpr.
tunmx
commented on July 17, 2024
感谢反馈，我会去找一台jetson nano尝试复现出问题
from hyperlpr.
Related Issues (20)
- c++版本 HOT 1
- 纯数字字母车牌的识别
- 多线程并发起来有些车牌无法识别到 HOT 3
- 角度多大会识别不了？ HOT 1
- 摩托车车牌无法识别 HOT 2
- 如何通过源码部署web服务 HOT 3
- Assessment of the difficulty in porting CPU architecture for HyperLPR
- 如何自定义数据集
- iOS Frameworks如何配置模型文件地址 HOT 4
- 有没有build_release_iOS_share.sh的脚本 HOT 1
- cpp编译依赖库123云盘分享，加速下载
- 牛逼的开源项目，感谢大佬们 HOT 1
- 用API和跑demo两个方法测试结果不一样
- When i import hyperlpr3 as lpr3 something error HOT 2
- 请问支持返回车牌的关键点位置吗？ HOT 2
- python 接口没有提供使用GPU的入参 HOT 2
- java.lang.UnsatisfiedLinkError: dlopen failed: library "libhyperlpr3.so" not found HOT 3
- v2版本能添加车牌颜色获取吗 HOT 1
- 黄牌双层识别错误
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from hyperlpr.