Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Error in CPU optimization #109

Open
isgursoy opened this issue Nov 21, 2022 · 2 comments
Open

Error in CPU optimization #109

isgursoy opened this issue Nov 21, 2022 · 2 comments
Labels
speedster Issue related to the Speedster App

Comments

@isgursoy
Copy link

21/11/2022 07:29:41 PM [ INFO ] Running Nebullvm optimization on CPU
21/11/2022 07:29:43 PM [ WARNING ] Missing Frameworks: tensorflow.
 Please install them to include them in the optimization pipeline.
21/11/2022 07:29:43 PM [ WARNING ] Missing Compilers: openvino.
 Please install them to include them in the optimization pipeline.
21/11/2022 07:29:43 PM [ WARNING ] Not enough data for splitting the DataManager. You should provide at least 100 data samples to allow a good split between train and test sets. Compression, calibration and precision checks will use the same data.
21/11/2022 07:29:44 PM [ INFO ] Converting the torch model to onnx
21/11/2022 07:30:29 PM [ INFO ] Benchmark performance of original model
21/11/2022 07:31:00 PM [ INFO ] Original model latency: 0.27246932983398436 sec/iter
21/11/2022 07:31:00 PM [ INFO ] Running pipeline: pytorch_pipeline
21/11/2022 07:31:00 PM [ INFO ] Running step: no_compression
21/11/2022 07:31:00 PM [ INFO ] Running step: optimizer_step
21/11/2022 07:31:00 PM [ INFO ] Optimizations: ('torchscript', 'deepsparse', 'intel_neural_compressor')
21/11/2022 07:31:00 PM [ INFO ] Optimizing output of compressor 1/1: no_compression
21/11/2022 07:31:00 PM [ INFO ] [ OPTIMIZER 1/3 ]
21/11/2022 07:31:00 PM [ INFO ] Optimizing with PytorchBackendOptimizer and q_type: None.
21/11/2022 07:31:29 PM [ INFO ] Optimized model latency: 0.23684096336364746 sec/iter
21/11/2022 07:31:29 PM [ INFO ] [ OPTIMIZER 2/3 ]
21/11/2022 07:31:29 PM [ INFO ] Optimizing with DeepSparseOptimizer and q_type: None.
DeepSparse Engine, Copyright 2021-present / Neuralmagic, Inc. version: 1.2.0 COMMUNITY EDITION | (45d54d49) (release) (optimized) (system=avx2, binary=avx2)
[7fcfbcaa8000 >ERROR< validate src/lib/engine/compute/primitive/stacked_gemm.cpp:104] Bias width does not match the number of output columns. Bias: 1,9604,1,1,1, B_matrix: 64,4,1,32,49
DeepSparse Engine, Copyright 2021-present / Neuralmagic, Inc. version: 1.2.0 (45d54d49) (release) (optimized) (system=avx2, binary=avx2)
Date: 11-21-2022 @ 19:32:22 +03
OS: Linux isgursoy-VORKE 5.15.0-54-generic #60-Ubuntu SMP Mon Nov 14 13:48:30 UTC 2022
Arch: x86_64
ISA: x86_64-avx2-32-64-16-64-xbyak
CPU: GenuineIntel
Vendor: Intel
Cores/sockets/threads: [4, 1, 8]
Available cores/sockets/threads: [4, 1, 8]
L1 cache size data/instruction: 32k/32k
L2 cache size: 0.25Mb
L3 cache size: 6Mb
Total memory: 17.5404G
Free memory: 1.51531G

Assertion at ./src/include/wand/engine/debug.hpp:233

Backtrace:
 0# wand::detail::abort_prefix(std::ostream&, char const*, char const*, int, bool, bool, unsigned long) in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
 1# 0x00007FCE7C9A05D4 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
 2# 0x00007FCE7C9A1092 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
 3# 0x00007FCE7C643993 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
 4# 0x00007FCE7C643CA2 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
 5# 0x00007FCE7C64A64A in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
 6# 0x00007FCE7C64A6C0 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
 7# 0x00007FCE7BA7A0F7 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
 8# 0x00007FCE7BA7A5EF in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
 9# 0x00007FCE7B9EEB35 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
10# 0x00007FCE7B9DE609 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
11# 0x00007FCE7B9DF8E1 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
12# 0x00007FCE7C144590 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
13# 0x00007FCE7C1499CE in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
14# 0x00007FCE7C14C527 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
15# 0x00007FCE7C14C954 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
16# 0x00007FCE7B9BC75E in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
17# 0x00007FCE7B9C434D in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
18# 0x00007FCE7B95C6ED in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
19# 0x00007FCE7B968E15 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libonnxruntime.so.1.12.0
20# deepsparse::ort_engine::init(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, int, std::shared_ptr<wand::parallel::scheduler_factory_t>) in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/libdeepsparse.so
21# 0x00007FCE7E6B4433 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/deepsparse_engine.so
22# 0x00007FCE7E6F06B9 in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/deepsparse_engine.so
23# 0x00007FCE7E6C1FFE in /opt/projects/akil/classification-engine/venv/lib/python3.10/site-packages/deepsparse/avx2/deepsparse_engine.so

Please email a copy of this stack trace and any additional information to: support@neuralmagic.com

Process finished with exit code 134 (interrupted by signal 6: SIGABRT)

pip git installation. Ubuntu 22.04.

@valeriosofi
Copy link
Collaborator

This is strange, I think that also here the cause is python 3.10, but it shouldn't crash like that. I'll investigate as soon as I can, for now you can set ignore_compilers=["deepsparse"] to avoid its usage.

@isgursoy
Copy link
Author

isgursoy commented Nov 22, 2022

Yes, crash is gone now. Lets have a cross reference under issue #103.

@francis-oss francis-oss added the speedster Issue related to the Speedster App label Jan 10, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
speedster Issue related to the Speedster App
Projects
None yet
Development

No branches or pull requests

3 participants