GithubHelp home page GithubHelp logo

axera-tech / pulsar-docs Goto Github PK

View Code? Open in Web Editor NEW
7.0 7.0 5.0 2.12 MB

The docs repository of Pulsar which is AXera's SoC AI toolchain. Such as AX630A, AX620A, AX620U

License: BSD 3-Clause "New" or "Revised" License

Makefile 20.06% Python 79.94%
ai compiler documentation toolchains

pulsar-docs's People

Contributors

bug1989 avatar

Stargazers

avatar avatar avatar avatar avatar avatar avatar

Watchers

avatar avatar

Forkers

bug1989 zihan987 zheqiushui mzpmzk yummy0929

pulsar-docs's Issues

Error when running basic ResNet18 export example

I am running:

pulsar run\
    resnet18_export_data/model/resnet18.joint resnet18_export_data/model/resnet18.onnx\
    --input resnet18_export_data/images/cat.jpg\
    --config resnet18_export_data/config/output_config.prototxt\
    --output_gt inference_results  # weird arg name

using the Resnet18 quick start data. However, I get an error:

[18 11:00:46 <frozen super_pulsar.model_executor>:266] DBG [pulsar build] File "<frozen super_pulsar.toolchain_wrappers.wrapper_hat_maker>", line 117, in get_io_modification_onnx
[18 11:00:46 <frozen super_pulsar.model_executor>:266] DBG [pulsar build] TypeError: get_modification_onnx() missing 1 required positional argument: 'meta_dict'
[18 11:00:46 <frozen super_pulsar.func_wrappers.wrapper_pulsar_run>:242] ERR failed loading resnet18_export_data/model/resnet18.onnx, skipping
[18 11:00:46 <frozen super_pulsar.func_wrappers.wrapper_pulsar_run>:244] DBG Traceback (most recent call last):
  File "<frozen super_pulsar.func_wrappers.wrapper_pulsar_run>", line 240, in main
  File "<frozen super_pulsar.func_wrappers.pulsar_run.utils>", line 31, in load_model
  File "<frozen super_pulsar.model_executor>", line 338, in __init__
  File "<frozen super_pulsar.model_executor>", line 271, in wrap_src_model_to_joint
RuntimeError: pulsar build returned 1

.joint get's exported. However, the ONNX model cannot be loaded and hence the inference comparison cannot be executed. It seems to lack some metadata. Any idea on how to fix this?

Version information of my toolchain:

root@0bed5bfd8f7b:/data# pulsar version
[W Context.cpp:69] Warning: torch.set_deterministic is in beta, and its design and  functionality may change in the future. (function operator())
0.6.1.20
07305a6

Run inference with `.joint` model

I want to run inference on my exported .joint model to be able to evaluate its INT8 COCO mAP performance. I only see one way of doing this:

pulsar run\
    my_model.joint\
    --input resnet18_export_data/images/cat.jpg\
    --output_gt inference_results

And then read the the generated .npy file. I want to avoid reading from file as it is a very time consuming operation. Is there a way of returning the results from the system call itself? Are you planning to implement this?

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo 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.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.