Cannot deploy `pt_pointpillars_nuscenes_40000_64_108G_2.5.zip` to KV260 #1116

ryo-takahashi-1206 · 2022-12-20T12:40:22Z

What I want to do:

I want to deploy pt_pointpillars_nuscenes_40000_64_108G_2.5.zip to my KV260

What I have done so far:

Build xilinx/vitis-ai-gpu:2.5.0.1260
- Use Vitis-AI.git/docker/docker_build_gpu.sh[469a4b7b]
- torch==1.10.1 was installed
Extend the above image basically according to pt_pointpillars_nuscenes_40000_64_108G_2.5.zip/README.md
- Use pt_pointpillars_nuscenes_40000_64_108G_2.5.zip/code/ rather than the upstream open-mmlab/mmdetection3d.git
Run pt_pointpillars_nuscenes_40000_64_108G_2.5.zip/run_quant.sh on the following environment
- GPU: NVIDIA Tesla V100
- Driver: 470.141.03
- CUDA Version: 11.4

What my issue is:

Crushed at the end of run_quant.sh (specifically, --dump_xmodel)
- LLVM ERROR: out of memory can be seen

python code/mmdetection3d/tools/quant.py float/hv_pointpillars_secfpn_sbn-all_4x4_2x_nus-3d_quant.py float/pointpillars-nus.pth --quant_mode test --quant_dir quantized --fast_finetune --dump_xmodel
Dumping xmodel...

[VAIQ_NOTE]: Loading NNDCT kernels...

[VAIQ_NOTE]: Quant config file is empty, use default quant configuration

[VAIQ_NOTE]: Quantization test process start up...

[VAIQ_NOTE]: =>Quant Module is in 'cpu'.

[VAIQ_NOTE]: =>Parsing MVXFasterRCNN_quant...

[VAIQ_NOTE]: Start to trace model...

[VAIQ_NOTE]: Finish tracing.

[VAIQ_NOTE]: Processing ops...

[VAIQ_WARN]: The quantizer recognize new op `index_put_inplace` as a float operator by default.

[VAIQ_WARN]: The quantizer recognize new op `equal` as a float operator by default.

[VAIQ_WARN]: The quantizer recognize new op `stack` as a float operator by default.
                                                                                         
[VAIQ_WARN]: The quantizer recognize new op `select` as a float operator by default.

[VAIQ_WARN]: The quantizer recognize new op `index` as a float operator by default.

[VAIQ_NOTE]: =>Doing weights equalization...

[VAIQ_NOTE]: =>Quantizable module is generated.(quantized/MVXFasterRCNN_quant.py)

[VAIQ_NOTE]: =>Get module with quantization.
                                                                                         
[VAIQ_NOTE]: =>Loading quant model parameters.(quantized/param.pth)                      [                                                  

[VAIQ_NOTE]: =>Converting to xmodel ...

[VAIQ_NOTE]: =>Dumping 'MVXFasterRCNN_quant_0'' checking data...

[VAIQ_WARN]: Only dump first output of multi-output node:'MVXFasterRCNN_quant::MVXFasterRCNN_quant/HardVFE_deploy_trans_input_quant[pts_voxel_encoder]/VFELayer_deploy_quant[vfe_layers]/ModuleList[0]/Max[max]/inputs.3(max)'.

[VAIQ_WARN]: Only dump first output of multi-output node:'MVXFasterRCNN_quant::MVXFasterRCNN_quant/HardVFE_deploy_trans_input_quant[pts_voxel_encoder]/VFELayer_deploy_quant[vfe_layers]/ModuleList[0]/Max[max]/inputs.3(max)'.

[VAIQ_NOTE]: =>Finsh dumping data.(quantized/deploy_check_data_int/MVXFasterRCNN_quant_0)

[VAIQ_NOTE]: =>Dumping 'MVXFasterRCNN_quant_1'' checking data...

[VAIQ_NOTE]: =>Finsh dumping data.(quantized/deploy_check_data_int/MVXFasterRCNN_quant_1)
LLVM ERROR: out of memory
./run_quant.sh: line 47: 10503 Aborted

Showing bt from the core dump, I could find abnormal new() happen under libxir.so as below:
- in this sense, LLVM ERROR: out of memory is reasonable

#0  0x00007f7dadc2ae87 in raise () from /lib/x86_64-linux-gnu/libc.so.6
#1  0x00007f7dadc2c7f1 in abort () from /lib/x86_64-linux-gnu/libc.so.6
#2  0x00007f7be6a68088 in llvm::report_bad_alloc_error(char const*, bool) ()
  from /home/vitis-ai-user/.local/lib/python3.7/site-packages/llvmlite/binding/libllvmlite.so
#3  0x00007f7da9f0bb97 in operator new (sz=140173364611777) at ../../../../libstdc++-v3/libsupc++/new_op.cc:55
#4  0x00007f7da9f6126f in std::__new_allocator<char>::allocate (this=<optimized out>, __n=<optimized out>)
   at /home/conda/feedstock_root/build_artifacts/gcc_compilers_1652324151713/work/build/x86_64-conda-linux-gnu/libstdc++-v3/src/c++11/alloc_traits.h:112
#5  std::allocator_traits<std::allocator<char> >::allocate (
   __a=<error reading variable: dwarf2_find_location_expression: Corrupted DWARF expression.>, __n=<optimized out>)
   at /home/conda/feedstock_root/build_artifacts/gcc_compilers_1652324151713/work/build/x86_64-conda-linux-gnu/libstdc++-v3/src/c++11/basic_string.tcc:464
#6  std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::_M_create (this=this@entry=0x7fff0b8e3800,
   __capacity=__capacity@entry=@0x7fff0b8e3288: 140173364611776, __old_capacity=<optimized out>)
   at /home/conda/feedstock_root/build_artifacts/gcc_compilers_1652324151713/work/build/x86_64-conda-linux-gnu/libstdc++-v3/src/c++11/basic_string.h:155
#7  0x00007f7da9f61472 in std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >::_M_assign (this=0x7fff0b8e3800,
   __str=...)
   at /home/conda/feedstock_root/build_artifacts/gcc_compilers_1652324151713/work/build/x86_64-conda-linux-gnu/libstdc++-v3/src/c++11/char_traits.h:1132
#8  0x00007f7c14e28d56 in std::__detail::_Compiler<std::__cxx11::regex_traits<char> >::_M_atom() ()
  from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/../../libxir.so.2
#9  0x00007f7c14e29279 in std::__detail::_Compiler<std::__cxx11::regex_traits<char> >::_M_alternative() ()
  from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/../../libxir.so.2
#10 0x00007f7c14e294cd in std::__detail::_Compiler<std::__cxx11::regex_traits<char> >::_M_disjunction() ()
  from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/../../libxir.so.2
#11 0x00007f7c14e29c26 in std::__detail::_Compiler<std::__cxx11::regex_traits<char> >::_Compiler(char const*, char const*, std::locale const&, std::regex_constants::syntax_option_type) () from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/../../libxir.so.2
#12 0x00007f7c14e2a602 in xir::DataType::DataType(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) ()
  from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/../../libxir.so.2
#13 0x00007f7c14e67cc4 in xir::OpImp::create_output_tensor_(xir::DataType const&) ()
  from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/../../libxir.so.2
#14 0x00007f7c14e6fa36 in xir::OpImp::OpImp(void*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unique_ptr<xir::Attrs, std::default_delete<xir::Attrs> >, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<xir::Op*, std::allocator<xir::Op*> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::vector<xir::Op*, std::allocator<xir::Op*> > > > > const&, std::unique_ptr<xir::Tensor, std::default_delete<xir::Tensor> >, xir::GraphImp*, xir::DataType const&) ()
  from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/../../libxir.so.2
#15 0x00007f7c14eb67fd in xir::GraphImp::add_op(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unique_ptr<xir::Attrs, std::default_delete<xir::Attrs> >, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<xir::Op*, std::allocator<xir::Op*> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::vector<xir::Op*, std::allocator<xir::Op*> > > > > const&, xir::DataType const&, xir::Subgraph*) () from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/../../libxir.so.2
#16 0x00007f7c14eafbc9 in xir::GraphImp::add_op(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::unique_ptr<xir::Attrs, std::default_delete<xir::Attrs> >, std::map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::vector<xir::Op*, std::allocator<xir::Op*> >, std::less<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, std::vector<xir::Op*, std::allocator<xir::Op*> > > > > const&, xir::Subgraph*) ()
  from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/../../libxir.so.2
#17 0x00007f7d73ff661a in ?? () from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/xir.cpython-37m-x86_64-linux-gnu.so
#18 0x00007f7d73fc7b97 in ?? () from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/xir.cpython-37m-x86_64-linux-gnu.so
#19 0x00007f7d74000c99 in ?? () from /opt/vitis_ai/conda/envs/vitis-ai-pytorch/lib/python3.7/site-packages/xir.cpython-37m-x86_64-linux-gnu.so
#20 0x000055f9932ad107 in _PyMethodDef_RawFastCallKeywords ()
   at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Objects/call.c:693
#21 0x000055f9932ad296 in _PyCFunction_FastCallKeywords (func=0x7f7d742cb870, args=<optimized out>, nargs=<optimized out>,
   kwnames=<optimized out>) at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Objects/call.c:732
#22 0x000055f9932ae269 in call_function () at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Python/ceval.c:4619
#23 0x000055f993324cba in _PyEval_EvalFrameDefault ()
   at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Python/ceval.c:3093
#24 0x000055f9932ab1d4 in PyEval_EvalFrameEx (throwflag=0, f=0x7f7bd7f74b30)
   at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Python/ceval.c:544
#25 function_code_fastcall (globals=0x7f7c1440b1e0, nargs=<optimized out>, args=<optimized out>, co=<optimized out>)
   at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Objects/call.c:283
#26 _PyFunction_FastCallKeywords () at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Objects/call.c:408
#27 0x000055f9932ae0d8 in call_function () at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Python/ceval.c:4616
#28 0x000055f9933208f1 in _PyEval_EvalFrameDefault ()
   at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Python/ceval.c:3110
#29 0x000055f993264ea2 in PyEval_EvalFrameEx (throwflag=0, f=0x7f7bdd3b3430)
   at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Python/ceval.c:3930
#30 _PyEval_EvalCodeWithName () at /home/conda/feedstock_root/build_artifacts/python_1635226054942/work/Python/ceval.c:3930

What I feel suspicious about:

My auto-generated MVXFasterRCNN_quant.py is slightly different from the archived pt_pointpillars_nuscenes_40000_64_108G_2.5.zip/quantized/MVXFasterRCNN_quant.py
- archived: seems like 2D
```
self.output_module_12 = self.module_12(end=[2147483647,2147483647], input=self.output_module_2, step=[1,1], start=[0,0])
```
- my artifact: seems like 1D
```
output_module_14 = self.module_14(input=output_module_3, dim=[1], start=[0], end=[9223372036854775807], step=[1])
```
- This difference could lead to abnormal new() like above ?

I'm sorry if my trouble is a very basic matter.
I would greatly appreciate any advice the community members could give me.

The text was updated successfully, but these errors were encountered:

ryo-takahashi-1206 · 2023-01-02T06:24:59Z

I found the above LLVM ERROR: out of memory happened during a runtime compilation of std::regex in src/Vitis-AI-Runtime/VART/xir/src/xir/util/data_type.cpp. As a result of eliminating std::regex in this file, I could confirmed the same error didn't happen again. I close this issue. I'd appreciate you if someone would bring more elegant solution into the upstream.

@hao

* [wego][examples] update release examples for vai-3.5 release. (#1025) Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * [examples][wego-tf2.x] remove erfnet from wego-tf2.x examples. (#1026) Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * [examples][wego_tf2.x] remove link for erfnet since it's not avaliable in vai-3.5 release. (#1027) Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * vai3.5 src update (#1028) Co-authored-by: qianglin-xlnx <linqiang@xilinx.com> * vai3.5 examples update (#1029) Co-authored-by: qianglin-xlnx <linqiang@xilinx.com> * vai3.5 board_setup update (#1030) Co-authored-by: qianglin-xlnx <linqiang@xilinx.com> * update links in README_DPUCVDX8G.md (#1031) Co-authored-by: xlinyang <xlinyang@xilinx.com> * [wego] add examples for pytorch 2.0 preview. (#1033) Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * Initial prototype of Github.IO documentation (#1034) * Update ONNX Runtime docs, dpu/README.md * Fix model_zoo/README.md Github.IO link, update Zoo license link in HTML docs * Fix Docker image naming convention and run commands * Fix Docker image naming convention and run commands * Update CUDA container test to align with release CUDA version * Fix Docker image naming convention and run commands * Fix broken Model Zoo license agreement link * Minor text updates to workflow HTML * Further optimization of install instructions * Further optimization of install instructions * Correct Model Zoo links in board_setup * Consolidate install_docker into install * Correct spelling mistake * Minor text updates to workflow * Minor text updates to workflow-model-zoo * Further updates to install instructions * Further updates to install instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Revise Alveo support, board_setup docs, Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Fix Model Zoo downloader.py typo * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix broken workflow link * Add V70 DPU description * Add V70 DPU description * Fix broken third-party link * Update EDD2020 dataset link * Updates for TOCTree structure * Major revisions to 3.0 docs * Rebrand quickstarts * Fix release notes links: * Initial 3.5 doc changes * Initial changes for 3.5 docs * Initial Doxygen integration test * Fixed Command in Install.rst * Fixed quickstart optimization links * Fixed quickstart optimization links * Update VAI_IDE.PNG * Delete VAI_IDE.PNG * Updated VAI_IDE.PNG Co-authored-by: Rhind <arhind@amd.com> * [fix] fix some build issue (#1032) Co-authored-by: shili <lishi@xilinx.com> * [fix] sync last 3 commit in vitis-ai-library (#1035) Co-authored-by: shili <lishi@xilinx.com> * [fix] python version 3.9->3.10 in op_reg.sh (#1037) * [fix] python version 3.9->3.10 in op_reg.sh * [fix] update some readme Co-authored-by: shili <lishi@xilinx.com> * [wego][fix] fix typos for wego pt2.0 preview example. (#1038) sync-up info: https://gitenterprise.xilinx.com/VitisAI/wego-release-examples/commit/a67a6a7580464c791da481c2125408616914dde6 Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * update vai 3.5 docker related (#1040) * update vai 3.5 docker related * update rocm installation section * update rocm installation section * update rocm installation section Co-authored-by: qiuyuny <qiuyuny@xilinx.com> * [wego][improvment] tune parameters for wego zendnn examples for better performance. (#1042) sync-up: https://gitenterprise.xilinx.com/VitisAI/wego-release-examples/commit/35f89cc8b372b5a54ca0a3e1e6e6333029092776 Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * [fix] update xmodel (#1041) * [fix] update xmodel * [fix] remove ocr samples Co-authored-by: shili <lishi@xilinx.com> * [wego] update pytorch examples with onboard support. (#1043) This new feature is implemented by @hao Zhuan, please refer to MR: https://gitenterprise.xilinx.com/VitisAI/wego-release-examples/pull/10 for more details. Co-authored-by: Hao Zhuan <haoz@xilinx.com> * update driver version for nvidia GPU docker requirements * add v70 setup flow and delete vck5000 setup folder * update vai_library samples * update vai_runtime examples * add Transformer examples link * update vai_library samples about V70 * update vai_runtime samples download link * add vck5000 board setup link * add OnBoard example * [onboard]demo format fix * [onboard]tiny format fix * add board_setup_v70.rst for bash * [fix] update readme (#1050) * [fix] update readme * [fix] multitaskv3 <-> multitask_v3 name change Co-authored-by: shili <lishi@xilinx.com> * Vai 3.5 update (#1052) * Update ONNX Runtime docs, dpu/README.md * Fix model_zoo/README.md Github.IO link, update Zoo license link in HTML docs * Fix Docker image naming convention and run commands * Fix Docker image naming convention and run commands * Update CUDA container test to align with release CUDA version * Fix Docker image naming convention and run commands * Fix broken Model Zoo license agreement link * Minor text updates to workflow HTML * Further optimization of install instructions * Further optimization of install instructions * Correct Model Zoo links in board_setup * Consolidate install_docker into install * Correct spelling mistake * Minor text updates to workflow * Minor text updates to workflow-model-zoo * Further updates to install instructions * Further updates to install instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Revise Alveo support, board_setup docs, Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Fix Model Zoo downloader.py typo * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix broken workflow link * Add V70 DPU description * Add V70 DPU description * Fix broken third-party link * Update EDD2020 dataset link * Updates for TOCTree structure * Major revisions to 3.0 docs * Rebrand quickstarts * Fix release notes links: * Initial 3.5 doc changes * Initial changes for 3.5 docs * Initial Doxygen integration test * Fixed Command in Install.rst * Fixed quickstart optimization links * Fixed quickstart optimization links * Update VAI_IDE.PNG * Delete VAI_IDE.PNG * Updated VAI_IDE.PNG * Merging Mpsoc quick start and setting up the target * MPSOC Quick Start Upload Test * Updated to new AMD theme * Update Sphinx theme * Updates to MPSoC quickstart * Updates to MPSoC quickstart * Updates to MPSoC quickstart * MPSOC Changes * New MPSOC + Quentons Changes * Remove Chapter 8 reference * Merging 3.0 changes to 3.5 * Fixed table format * Minor updates * Fixed MPSOC.rst * Various 3.5 documentation updates * Various 3.5 documentation updates * First pass V70 quickstart Co-authored-by: Rhind <arhind@amd.com> Co-authored-by: Aidan Rhind <aidan.rhind@xilinx.com> * update onnx runtime installation and link error fix (#1053) Co-authored-by: qianglin-xlnx <linqiang@xilinx.com> * update README_DPUCV2DX8G.md (#1054) * update pytorch quantizer code to 3.5 (#1055) Co-authored-by: niuxj <niuxj@xilinx.com> * update Setup V70 Accelerator Card link * update model.yaml * update the model name * [wego][fix] fix issue fired in jira: https://jira.xilinx.com/browse/VAI-2927?src=confmacro (#1058) sync-up: https://gitenterprise.xilinx.com/VitisAI/wego-release-examples/commit/c2305da0ddc714c0e248e9d5252cd7359dc60982 Co-authored-by: Xie, ChuanLiang <ChuanLiang.Xie@amd.com> Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * [vai_library][apps] update vek280 demos (#1059) * update OnBoard model path name * [fix] update unet2d readme (#1061) Co-authored-by: shili <lishi@xilinx.com> * [wego][fix] fix issue fired in jira: https://jira.xilinx.com/browse/VAI-2926 (#1062) Sync-up: https://gitenterprise.xilinx.com/VitisAI/wego-release-examples/commit/9e570fa4be75c87cf8a67e41277cabdcd3986878 Co-authored-by: Sen Wang <Sen.Wang1@amd.com> * Add vai_optimizer * update modelzoo readme * update readme (#1039) * update readme * fix * update OnBoard readme * [fix] update yolov578 readme (#1065) Co-authored-by: shili <lishi@xilinx.com> * update v70 quick start steps * update the vck5000 and v70 quick start steps * update for 3.5 release * update V70 setup link * update the V70 DPU xclbin config * [fix] fix readme for accuracy test (#1069) Co-authored-by: shili <lishi@xilinx.com> * [wego][fix] remove stackoverflow link to avoid legal issue. (#1070) Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * [fix] avoid onnx sample accuracy hang (#1072) Co-authored-by: shili <lishi@xilinx.com> * Update README.md * Delete README.md.old * [fix] onnx model poinptillars readme update; dynamic batch support update (#1074) Co-authored-by: shili <lishi@xilinx.com> * Update release_notes.rst: add Optimizer and PyTorch CNN quantizer * [fix] vai_profiler README was moved to github.io (#1076) Co-authored-by: LiYunzhi <liyunzhi@xilinx.com> * Vai 3.5 update (#1075) * Update ONNX Runtime docs, dpu/README.md * Fix model_zoo/README.md Github.IO link, update Zoo license link in HTML docs * Fix Docker image naming convention and run commands * Fix Docker image naming convention and run commands * Update CUDA container test to align with release CUDA version * Fix Docker image naming convention and run commands * Fix broken Model Zoo license agreement link * Minor text updates to workflow HTML * Further optimization of install instructions * Further optimization of install instructions * Correct Model Zoo links in board_setup * Consolidate install_docker into install * Correct spelling mistake * Minor text updates to workflow * Minor text updates to workflow-model-zoo * Further updates to install instructions * Further updates to install instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Revise Alveo support, board_setup docs, Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Fix Model Zoo downloader.py typo * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix broken workflow link * Add V70 DPU description * Add V70 DPU description * Fix broken third-party link * Update EDD2020 dataset link * Updates for TOCTree structure * Major revisions to 3.0 docs * Rebrand quickstarts * Fix release notes links: * Initial 3.5 doc changes * Initial changes for 3.5 docs * Initial Doxygen integration test * Fixed Command in Install.rst * Fixed quickstart optimization links * Fixed quickstart optimization links * Update VAI_IDE.PNG * Delete VAI_IDE.PNG * Updated VAI_IDE.PNG * Merging Mpsoc quick start and setting up the target * MPSOC Quick Start Upload Test * Updated to new AMD theme * Update Sphinx theme * Updates to MPSoC quickstart * Updates to MPSoC quickstart * Updates to MPSoC quickstart * MPSOC Changes * New MPSOC + Quentons Changes * Remove Chapter 8 reference * Merging 3.0 changes to 3.5 * Fixed table format * Minor updates * Fixed MPSOC.rst * Various 3.5 documentation updates * Various 3.5 documentation updates * First pass V70 quickstart * GithubIO & repo updates for 3.5 * Remove RNN quantizer and Transformer references * Updated to remove legal keywords * Update V70 Quickstart, remove VCK5000 doc * Synced latest changes Co-authored-by: Rhind <arhind@amd.com> Co-authored-by: Aidan Rhind <aidan.rhind@xilinx.com> * Update release_notes.rst * [fix] port coverity fix from VaiLib (#1077) * [fix] port coverity fix from VaiLib * [fix] remove useless url in comments Co-authored-by: shili <lishi@xilinx.com> * update release_notes and vek280 board setup (#1078) * update model list * update OnBoard readme * update v70 link * update onboard link * update AIlibrary sample link * update vai runtime link * update v70 setup link * Update LICENSE (#1073) * Update LICENSE * Update LICENSE * [Example][OnBoard]license updated * [examples][vai_library] update vek280 4video demo (#1081) * [optimizer] Add license header to tf2 mnist example * Vai 3.5 update (#1084) * [vai_q_tensorflow1.x] remove 3.0 folder * [vai_q_tensorflow1.x] add 3.5 folder, from dev and 3.5 branch commit 4f8667a * [vai_q_tensorflow2.x] remove 3.0 folder * [vai_q_tensorflow2.x] add 3.5 folder, from dev and 3.5 branch commit f8021be * Add vai_q_onnx (#1083) * Vai 3.5 update recipes (#1082) * src:vai_petalinux_recipes update recipes-vitis-ai to 3.5 1. update recipes-vitis-ai to 3.5 2. add recipes-devtools and recipes-support for vai35 * src:vai_petalinux_recipes update bsp.cfg for vaitracer * Update README.md * add custom op V70 compiled model * add apu install command * [fix] coverity fix: port from VaiLib (#1087) Co-authored-by: shili <lishi@xilinx.com> * Additional 3.5 updates (#1085) * Update ONNX Runtime docs, dpu/README.md * Fix model_zoo/README.md Github.IO link, update Zoo license link in HTML docs * Fix Docker image naming convention and run commands * Fix Docker image naming convention and run commands * Update CUDA container test to align with release CUDA version * Fix Docker image naming convention and run commands * Fix broken Model Zoo license agreement link * Minor text updates to workflow HTML * Further optimization of install instructions * Further optimization of install instructions * Correct Model Zoo links in board_setup * Consolidate install_docker into install * Correct spelling mistake * Minor text updates to workflow * Minor text updates to workflow-model-zoo * Further updates to install instructions * Further updates to install instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Revise Alveo support, board_setup docs, Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Fix Model Zoo downloader.py typo * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix broken workflow link * Add V70 DPU description * Add V70 DPU description * Fix broken third-party link * Update EDD2020 dataset link * Updates for TOCTree structure * Major revisions to 3.0 docs * Rebrand quickstarts * Fix release notes links: * Initial 3.5 doc changes * Initial changes for 3.5 docs * Initial Doxygen integration test * Fixed Command in Install.rst * Fixed quickstart optimization links * Fixed quickstart optimization links * Update VAI_IDE.PNG * Delete VAI_IDE.PNG * Updated VAI_IDE.PNG * Merging Mpsoc quick start and setting up the target * MPSOC Quick Start Upload Test * Updated to new AMD theme * Update Sphinx theme * Updates to MPSoC quickstart * Updates to MPSoC quickstart * Updates to MPSoC quickstart * MPSOC Changes * New MPSOC + Quentons Changes * Remove Chapter 8 reference * Merging 3.0 changes to 3.5 * Fixed table format * Minor updates * Fixed MPSOC.rst * Various 3.5 documentation updates * Various 3.5 documentation updates * First pass V70 quickstart * GithubIO & repo updates for 3.5 * Remove RNN quantizer and Transformer references * Updated to remove legal keywords * Update V70 Quickstart, remove VCK5000 doc * Synced latest changes * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Remove historic ONNX runtime EP docs * Update custom_target.rst * Update custom_target.rst * Fix v70.rst formatting Co-authored-by: Rhind <arhind@amd.com> Co-authored-by: Aidan Rhind <aidan.rhind@xilinx.com> * [fix][yolov8] fix yolov8 run error with large image (#1088) * Vai 3.5 update add tensorflow-onnx (#1090) * [vai_q_tensorflow1.x] remove 3.0 folder * [vai_q_tensorflow1.x] add 3.5 folder, from dev and 3.5 branch commit 4f8667a * [vai_q_tensorflow2.x] remove 3.0 folder * [vai_q_tensorflow2.x] add 3.5 folder, from dev and 3.5 branch commit f8021be * [tensorflow-onnx] open source for 3.5, from dev and 3.5 branch commit 78637ee * [doc][xv2] updated for xv2 changelog (#1095) * add XNNC (#1092) Co-authored-by: Zhenzhen Ding <zhenzhen.ding@xilinx.com> * [feat] add examples for vai_q_tensorflow2 (#1091) new file: __init__.py new file: candidate_layer_quantize.py new file: custom_quantize_strategy.json new file: export_onnx.py new file: fashion_mnist_bfp/main.py new file: fashion_mnist_bfp/net.py new file: fashion_mnist_bfp/run.sh new file: mnist_cnn_convert_datatype.py new file: mnist_cnn_inspect.py new file: mnist_cnn_ptq.py new file: mnist_cnn_ptq_config_quantize_strategy.py new file: mnist_cnn_ptq_custom_layer.py new file: mnist_cnn_ptq_fs.py new file: mnist_cnn_ptq_gpu.py new file: mnist_cnn_ptq_mix_precision.py new file: mnist_cnn_ptq_unsigned_quantize.py new file: mnist_cnn_qat.py new file: mnist_cnn_quantize_with_xcompiler.py new file: my_quantize_strategy.json new file: transformers_bert_qa/code/run_qa.py new file: transformers_bert_qa/code/run_qa_quantize.py new file: transformers_bert_qa/code/utils_qa.py new file: transformers_bert_qa/readme.md new file: transformers_bert_qa/requirements.txt new file: transformers_bert_qa/run_eval.sh new file: transformers_bert_qa/run_qatrain.sh new file: transformers_bert_qa/run_quantize.sh new file: transformers_bert_qa/run_quantize_eval.sh new file: transformers_bert_qa/run_train.sh Co-authored-by: Zhenzhen Ding <zhenzhen.ding@xilinx.com> * Vai 3.5 update (#1093) * [bugfix]:fix the tf1 onnx output format error * [bugfix]: fix the link in the file * [bugfix]:fix the numpy and onnx version in requirements.txt and setup.py * [bugfix]: fix the convert datatype & float scale quantizer model convert to onnx output format Co-authored-by: Jiangyong Ren <jiangyon@amd.com> * [wego][fix] update the instructions for VCK5000 * [WeGO][Fix]: Fix the computing formula of FPS * [feat]initially add vai_benchmark tool (#1099) * [feat]initial add vai_benchmark tool * remove author name and update README * [fix]add License Co-authored-by: Yanjun Zhang <yanjunz@xilinx.com> * add recipes for vai-benchmark * release_notes.rst: update PyTorch CNN quantizer * sync vai_q_pytorch 3.5 code (#1100) Co-authored-by: niuxj <niuxj@xilinx.com> * update docker recipe for 3.5 release (#1101) Co-authored-by: qiuyuny <qiuyuny@xilinx.com> * Vai 3.5 update (#1103) * [bugfix]:fix the tf1 onnx output format error * [bugfix]: fix the link in the file * [bugfix]:fix the numpy and onnx version in requirements.txt and setup.py * [bugfix]: fix the convert datatype & float scale quantizer model convert to onnx output format * [bugfix]:fix the bias_node should be biasadd_node * [bugfix] modify usage of README.md Co-authored-by: Jiangyong Ren <jiangyon@amd.com> * [examples][vai_q_tensorflow] add tf1 quantizer examples (#1102) * add XNNC * [examples][vai_q_tensorflow] add tf1 quantizer examples Co-authored-by: Zhenzhen Ding <zhenzhen.ding@xilinx.com> * update model list * update model list * [fix] synchronize code from Vitis-AI-Library (#1108) Co-authored-by: shili <lishi@xilinx.com> * remove some repo from permissive License list * update recipes-vitis-ai with vitis-staging mirror * recipe of protobuf static lib: clean * [fix] Recover vai profiler files (#1111) Co-authored-by: jileil <jilei.liu@xilinx.com> * delete src/vai_petalinux_recipes/recipes-devtools/ * update recipes for kernel * update recipe for kernel * Update vai_petalinux_recipes README * Update vai_petalinux_recip README * update vai_petalinux_recipes README * update vai_petalinux_recipes README * Update vai_petalinux_recipes xir: not use protobuf static * [feat]add vai_benchmark description (#1112) Co-authored-by: Yanjun Zhang <yanjunz@xilinx.com> * [fix] update vek280 model yaml (#1116) Co-authored-by: shili <lishi@xilinx.com> * Adjust vai_petalinux recipes structure * rename vai-benchmark.bb to vai-benchmark_3.5.bb * Vai 3.5 update (#1117) * Update ONNX Runtime docs, dpu/README.md * Fix model_zoo/README.md Github.IO link, update Zoo license link in HTML docs * Fix Docker image naming convention and run commands * Fix Docker image naming convention and run commands * Update CUDA container test to align with release CUDA version * Fix Docker image naming convention and run commands * Fix broken Model Zoo license agreement link * Minor text updates to workflow HTML * Further optimization of install instructions * Further optimization of install instructions * Correct Model Zoo links in board_setup * Consolidate install_docker into install * Correct spelling mistake * Minor text updates to workflow * Minor text updates to workflow-model-zoo * Further updates to install instructions * Further updates to install instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Revise Alveo support, board_setup docs, Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Fix Model Zoo downloader.py typo * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix broken workflow link * Add V70 DPU description * Add V70 DPU description * Fix broken third-party link * Update EDD2020 dataset link * Updates for TOCTree structure * Major revisions to 3.0 docs * Rebrand quickstarts * Fix release notes links: * Initial 3.5 doc changes * Initial changes for 3.5 docs * Initial Doxygen integration test * Fixed Command in Install.rst * Fixed quickstart optimization links * Fixed quickstart optimization links * Update VAI_IDE.PNG * Delete VAI_IDE.PNG * Updated VAI_IDE.PNG * Merging Mpsoc quick start and setting up the target * MPSOC Quick Start Upload Test * Updated to new AMD theme * Update Sphinx theme * Updates to MPSoC quickstart * Updates to MPSoC quickstart * Updates to MPSoC quickstart * MPSOC Changes * New MPSOC + Quentons Changes * Remove Chapter 8 reference * Merging 3.0 changes to 3.5 * Fixed table format * Minor updates * Fixed MPSOC.rst * Various 3.5 documentation updates * Various 3.5 documentation updates * First pass V70 quickstart * GithubIO & repo updates for 3.5 * Remove RNN quantizer and Transformer references * Updated to remove legal keywords * Update V70 Quickstart, remove VCK5000 doc * Synced latest changes * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Remove historic ONNX runtime EP docs * Update custom_target.rst * Update custom_target.rst * Fix v70.rst formatting * Update VEK280 Quickstart, Linux kernel version listings * Update VEK280 Quickstart, Linux kernel version listings * VEK280 + V70 Quickstarts * Updates to V70 and Model Zoo * Add Wallace.jpg * Merge changes across all Quickstarts * update for v70 quick start * Minor updates to Quickstarts * Editorial updates for 3.5 release Co-authored-by: Rhind <arhind@amd.com> Co-authored-by: Aidan Rhind <aidan.rhind@xilinx.com> Co-authored-by: suzhan <zhans@xilinx.com> * [wego] remove yolov5 PTQ example due to AGPL license issue. (#1118) Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * [fix][vai_library] remove yolov8 onnx example (#1120) * [fix] add 'examples' in default EXAMPLE_PATH (#1121) Co-authored-by: Yanjun Zhang <yanjunz@xilinx.com> * [fix] Correct vaitrace run path (#1119) Co-authored-by: jileil <jilei.liu@xilinx.com> * [fix] readme update due to issues from email (#1122) Co-authored-by: shili <lishi@xilinx.com> * [bugfix]:fix the pruned RCAN model quantization (#1107) * [bugfix]:fix the pruned RCAN model quantization * [vai_q_tensorflow][examples] add tf1 quantizer ptq example Co-authored-by: Jiangyong Ren <jiangyon@amd.com> * Update onnx version in requirements.txt and update the example (#1110) * Add vai_q_onnx * Update onnx version and example * Update onnx version and example * [examples][vai_q_tensorflow2] update README.md and example (#1115) Co-authored-by: Zhenzhen Ding <zhenzhen.ding@xilinx.com> * src/vai_petalinux_recipes/README.md: add packages version info * src/vai_petalinux_recipes/README.md: fix info * src/vai_petalinux_recipes/README.md: update * update model.yaml * remove vx2602 and ve2202 (#1123) * Update release_notes.rst * Vai 3.5 update (#1124) * Update ONNX Runtime docs, dpu/README.md * Fix model_zoo/README.md Github.IO link, update Zoo license link in HTML docs * Fix Docker image naming convention and run commands * Fix Docker image naming convention and run commands * Update CUDA container test to align with release CUDA version * Fix Docker image naming convention and run commands * Fix broken Model Zoo license agreement link * Minor text updates to workflow HTML * Further optimization of install instructions * Further optimization of install instructions * Correct Model Zoo links in board_setup * Consolidate install_docker into install * Correct spelling mistake * Minor text updates to workflow * Minor text updates to workflow-model-zoo * Further updates to install instructions * Further updates to install instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Revise Alveo support, board_setup docs, Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Fix Model Zoo downloader.py typo * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix broken workflow link * Add V70 DPU description * Add V70 DPU description * Fix broken third-party link * Update EDD2020 dataset link * Updates for TOCTree structure * Major revisions to 3.0 docs * Rebrand quickstarts * Fix release notes links: * Initial 3.5 doc changes * Initial changes for 3.5 docs * Initial Doxygen integration test * Fixed Command in Install.rst * Fixed quickstart optimization links * Fixed quickstart optimization links * Update VAI_IDE.PNG * Delete VAI_IDE.PNG * Updated VAI_IDE.PNG * Merging Mpsoc quick start and setting up the target * MPSOC Quick Start Upload Test * Updated to new AMD theme * Update Sphinx theme * Updates to MPSoC quickstart * Updates to MPSoC quickstart * Updates to MPSoC quickstart * MPSOC Changes * New MPSOC + Quentons Changes * Remove Chapter 8 reference * Merging 3.0 changes to 3.5 * Fixed table format * Minor updates * Fixed MPSOC.rst * Various 3.5 documentation updates * Various 3.5 documentation updates * First pass V70 quickstart * GithubIO & repo updates for 3.5 * Remove RNN quantizer and Transformer references * Updated to remove legal keywords * Update V70 Quickstart, remove VCK5000 doc * Synced latest changes * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Remove historic ONNX runtime EP docs * Update custom_target.rst * Update custom_target.rst * Fix v70.rst formatting * Update VEK280 Quickstart, Linux kernel version listings * Update VEK280 Quickstart, Linux kernel version listings * VEK280 + V70 Quickstarts * Updates to V70 and Model Zoo * Add Wallace.jpg * Merge changes across all Quickstarts * update for v70 quick start * Minor updates to Quickstarts * Editorial updates for 3.5 release * update v70 quickstart command * update v70 quickstart command * Update Linux Devicetree docs * Add Linux CMA Topic * WSL Docker Links Added * Doxygen and DPU nomenclature updates * Update Online Install * update v70 quick start command * Minor edits to Library online install * Remove VE2202/2602 from release notes * Fix Alveo V70 image * Fix Alveo targets discontinued statement Co-authored-by: Rhind <arhind@amd.com> Co-authored-by: Aidan Rhind <aidan.rhind@xilinx.com> Co-authored-by: suzhan <zhans@xilinx.com> * sync vaiq pytorch code and correct example directory (#1125) Co-authored-by: niuxj <niuxj@xilinx.com> * delete board_setup_v70.rst * [wego] add release notes for wego part. (#1127) Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> * update release note about V70 vitis version * Update src/vai_petalinux_recipes/README.md * Vai 3.5 update (#1129) * add XNNC * fix bugs in example and change default onnx_opset_version from 11 to 13 Co-authored-by: Zhenzhen Ding <zhenzhen.ding@xilinx.com> * Vai 3.5 update (#1130) * Update ONNX Runtime docs, dpu/README.md * Fix model_zoo/README.md Github.IO link, update Zoo license link in HTML docs * Fix Docker image naming convention and run commands * Fix Docker image naming convention and run commands * Update CUDA container test to align with release CUDA version * Fix Docker image naming convention and run commands * Fix broken Model Zoo license agreement link * Minor text updates to workflow HTML * Further optimization of install instructions * Further optimization of install instructions * Correct Model Zoo links in board_setup * Consolidate install_docker into install * Correct spelling mistake * Minor text updates to workflow * Minor text updates to workflow-model-zoo * Further updates to install instructions * Further updates to install instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Revise Alveo support, board_setup docs, Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Fix Model Zoo downloader.py typo * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix broken workflow link * Add V70 DPU description * Add V70 DPU description * Fix broken third-party link * Update EDD2020 dataset link * Updates for TOCTree structure * Major revisions to 3.0 docs * Rebrand quickstarts * Fix release notes links: * Initial 3.5 doc changes * Initial changes for 3.5 docs * Initial Doxygen integration test * Fixed Command in Install.rst * Fixed quickstart optimization links * Fixed quickstart optimization links * Update VAI_IDE.PNG * Delete VAI_IDE.PNG * Updated VAI_IDE.PNG * Merging Mpsoc quick start and setting up the target * MPSOC Quick Start Upload Test * Updated to new AMD theme * Update Sphinx theme * Updates to MPSoC quickstart * Updates to MPSoC quickstart * Updates to MPSoC quickstart * MPSOC Changes * New MPSOC + Quentons Changes * Remove Chapter 8 reference * Merging 3.0 changes to 3.5 * Fixed table format * Minor updates * Fixed MPSOC.rst * Various 3.5 documentation updates * Various 3.5 documentation updates * First pass V70 quickstart * GithubIO & repo updates for 3.5 * Remove RNN quantizer and Transformer references * Updated to remove legal keywords * Update V70 Quickstart, remove VCK5000 doc * Synced latest changes * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Remove historic ONNX runtime EP docs * Update custom_target.rst * Update custom_target.rst * Fix v70.rst formatting * Update VEK280 Quickstart, Linux kernel version listings * Update VEK280 Quickstart, Linux kernel version listings * VEK280 + V70 Quickstarts * Updates to V70 and Model Zoo * Add Wallace.jpg * Merge changes across all Quickstarts * update for v70 quick start * Minor updates to Quickstarts * Editorial updates for 3.5 release * update v70 quickstart command * update v70 quickstart command * Update Linux Devicetree docs * Add Linux CMA Topic * WSL Docker Links Added * Doxygen and DPU nomenclature updates * Update Online Install * update v70 quick start command * Minor edits to Library online install * Remove VE2202/2602 from release notes * Fix Alveo V70 image * Fix Alveo targets discontinued statement * Update Online Library Install * Update Online Library Install * Correct dumb number mistake * Correct dumb number mistake * Clean HTML build * Additional Editorial Changes * Restructure index.rst, convert VEK280 ref design to RST * Update Model Zoo to 3.5 * Fix minor formatting errors Co-authored-by: Rhind <arhind@amd.com> Co-authored-by: Aidan Rhind <aidan.rhind@xilinx.com> Co-authored-by: suzhan <zhans@xilinx.com> * vai 3.5 release (#1131) Co-authored-by: qiuyuny <qiuyuny@xilinx.com> * remove invalid links * [Release][bugfix]change v70 setup url to rel path (#1133) * update vart source to the latest (#1134) Co-authored-by: qianglin-xlnx <linqiang@xilinx.com> * Final updates for 3.5 public release (#1136) * Update ONNX Runtime docs, dpu/README.md * Fix model_zoo/README.md Github.IO link, update Zoo license link in HTML docs * Fix Docker image naming convention and run commands * Fix Docker image naming convention and run commands * Update CUDA container test to align with release CUDA version * Fix Docker image naming convention and run commands * Fix broken Model Zoo license agreement link * Minor text updates to workflow HTML * Further optimization of install instructions * Further optimization of install instructions * Correct Model Zoo links in board_setup * Consolidate install_docker into install * Correct spelling mistake * Minor text updates to workflow * Minor text updates to workflow-model-zoo * Further updates to install instructions * Further updates to install instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Cleanup VCK5000 setup instructions * Revise Alveo support, board_setup docs, Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Revise Alveo support / unify board_setup docs / update Model Zoo readme * Fix Model Zoo downloader.py typo * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Add Github.IO 404.html page * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix VCK5000 board_setup doc error * Fix broken workflow link * Add V70 DPU description * Add V70 DPU description * Fix broken third-party link * Update EDD2020 dataset link * Updates for TOCTree structure * Major revisions to 3.0 docs * Rebrand quickstarts * Fix release notes links: * Initial 3.5 doc changes * Initial changes for 3.5 docs * Initial Doxygen integration test * Fixed Command in Install.rst * Fixed quickstart optimization links * Fixed quickstart optimization links * Update VAI_IDE.PNG * Delete VAI_IDE.PNG * Updated VAI_IDE.PNG * Merging Mpsoc quick start and setting up the target * MPSOC Quick Start Upload Test * Updated to new AMD theme * Update Sphinx theme * Updates to MPSoC quickstart * Updates to MPSoC quickstart * Updates to MPSoC quickstart * MPSOC Changes * New MPSOC + Quentons Changes * Remove Chapter 8 reference * Merging 3.0 changes to 3.5 * Fixed table format * Minor updates * Fixed MPSOC.rst * Various 3.5 documentation updates * Various 3.5 documentation updates * First pass V70 quickstart * GithubIO & repo updates for 3.5 * Remove RNN quantizer and Transformer references * Updated to remove legal keywords * Update V70 Quickstart, remove VCK5000 doc * Synced latest changes * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Update Github to GithubIO inter-linkage * Remove historic ONNX runtime EP docs * Update custom_target.rst * Update custom_target.rst * Fix v70.rst formatting * Update VEK280 Quickstart, Linux kernel version listings * Update VEK280 Quickstart, Linux kernel version listings * VEK280 + V70 Quickstarts * Updates to V70 and Model Zoo * Add Wallace.jpg * Merge changes across all Quickstarts * update for v70 quick start * Minor updates to Quickstarts * Editorial updates for 3.5 release * update v70 quickstart command * update v70 quickstart command * Update Linux Devicetree docs * Add Linux CMA Topic * WSL Docker Links Added * Doxygen and DPU nomenclature updates * Update Online Install * update v70 quick start command * Minor edits to Library online install * Remove VE2202/2602 from release notes * Fix Alveo V70 image * Fix Alveo targets discontinued statement * Update Online Library Install * Update Online Library Install * Correct dumb number mistake * Correct dumb number mistake * Clean HTML build * Additional Editorial Changes * Restructure index.rst, convert VEK280 ref design to RST * Update Model Zoo to 3.5 * Fix minor formatting errors * Updates to Linux CMA, Release Notes, etc * Update VEK280 notes * new vek280 graphic, format fixing, resnet50_pt-=> resnet50 for example Co-authored-by: Rhind <arhind@amd.com> Co-authored-by: Aidan Rhind <aidan.rhind@xilinx.com> Co-authored-by: suzhan <zhans@xilinx.com> * [fix] fix tf2_2dunet yaml (#1139) Co-authored-by: shili <lishi@xilinx.com> Co-authored-by: Tianping Li <tianping.li@xilinx.com> Co-authored-by: Tianping Li <tianping@xcogpuvai02.xilinx.com> Co-authored-by: Qiang Lin <qiang.lin@xilinx.com> Co-authored-by: qianglin-xlnx <linqiang@xilinx.com> Co-authored-by: Xinlin Yang <xinlin.yang@xilinx.com> Co-authored-by: xlinyang <xlinyang@xilinx.com> Co-authored-by: Quenton Hall <quentonh@amd.com> Co-authored-by: Rhind <arhind@amd.com> Co-authored-by: Li Shi <li.shi@xilinx.com> Co-authored-by: shili <lishi@xilinx.com> Co-authored-by: qiuyuny <qiuyuny@xilinx.com> Co-authored-by: Hao Zhuan <haoz@xilinx.com> Co-authored-by: suzhan <zhans@xilinx.com> Co-authored-by: Zhan Su <zhan.su@amd.com> Co-authored-by: Aidan Rhind <aidan.rhind@xilinx.com> Co-authored-by: Peiyue Han <allanhan@xilinx.com> Co-authored-by: Xinjun Niu <xinjun.niu@xilinx.com> Co-authored-by: niuxj <niuxj@xilinx.com> Co-authored-by: wangxd <wangxd@xilinx.com> Co-authored-by: Xie, ChuanLiang <ChuanLiang.Xie@amd.com> Co-authored-by: Yu Zhou <yu.zhou@xilinx.com> Co-authored-by: Sen Wang <Sen.Wang1@amd.com> Co-authored-by: yuwang <yu.w@amd.com> Co-authored-by: Runfeng Wang <runfeng.wang@xilinx.com> Co-authored-by: Xiaodong Wang <xiaodong.wang@xilinx.com> Co-authored-by: Yunzhi Li <yunzhi.li@xilinx.com> Co-authored-by: LiYunzhi <liyunzhi@xilinx.com> Co-authored-by: GengXin Wu <gengxin.wu@xilinx.com> Co-authored-by: Chao Li <chao.li@xilinx.com> Co-authored-by: Yangyang Liang <ylia@xilinx.com> Co-authored-by: Jason Jia <xijie.jia@xilinx.com> Co-authored-by: Zhenzhen Ding <zhenzhen.ding@amd.com> Co-authored-by: Zhenzhen Ding <zhenzhen.ding@xilinx.com> Co-authored-by: Jiangyong Ren <jiangyong.ren@amd.com> Co-authored-by: Jiangyong Ren <jiangyon@amd.com> Co-authored-by: chuanliang <chuanlia@xilinx.com> Co-authored-by: chaow <chaow@amd.com> Co-authored-by: Yanjun Zhang <yanjun.zhang@xilinx.com> Co-authored-by: Yanjun Zhang <yanjunz@xilinx.com> Co-authored-by: Tianfang Meng <tianfang@amd.com> Co-authored-by: jianxuw <jianxuw@amd.com> Co-authored-by: JiLei Liu <jilei.liu@amd.com> Co-authored-by: jileil <jilei.liu@xilinx.com> Co-authored-by: Hao Zhuan <hao.zhuan@xilinx.com>

ryo-takahashi-1206 closed this as completed Jan 2, 2023

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Cannot deploy `pt_pointpillars_nuscenes_40000_64_108G_2.5.zip` to KV260 #1116

Cannot deploy `pt_pointpillars_nuscenes_40000_64_108G_2.5.zip` to KV260 #1116

ryo-takahashi-1206 commented Dec 20, 2022 •

edited

Loading

ryo-takahashi-1206 commented Jan 2, 2023

Cannot deploy pt_pointpillars_nuscenes_40000_64_108G_2.5.zip to KV260 #1116

Cannot deploy pt_pointpillars_nuscenes_40000_64_108G_2.5.zip to KV260 #1116

Comments

ryo-takahashi-1206 commented Dec 20, 2022 • edited Loading

What I want to do:

What I have done so far:

What my issue is:

What I feel suspicious about:

ryo-takahashi-1206 commented Jan 2, 2023

Cannot deploy `pt_pointpillars_nuscenes_40000_64_108G_2.5.zip` to KV260 #1116

Cannot deploy `pt_pointpillars_nuscenes_40000_64_108G_2.5.zip` to KV260 #1116

ryo-takahashi-1206 commented Dec 20, 2022 •

edited

Loading