davidmallasen / livechess2fen Goto Github PK
View Code? Open in Web Editor NEWPredict live chess games into FEN notation.
Home Page: https://arxiv.org/abs/2012.06858
License: GNU Affero General Public License v3.0
Predict live chess games into FEN notation.
Home Page: https://arxiv.org/abs/2012.06858
License: GNU Affero General Public License v3.0
Could the README file include all the library version for the framework dependencies
When #23 has been merged (thus resolving #17), the magic numbers (4
in this case) used here:
LiveChess2FEN/lc2fen/infer_pieces.py
Line 291 in 8a2bc33
infer_pieces.py
should be changed accordingly.I am relatively new to Jetson Nano, and I wanted to install your chess board recognition algorithm. I think I followed the steps pretty well, and in the onnx-tensorrt installation, when I run the make command in the build folder, I get the following error:
CMakeFiles/nvonnxparser.dir/build.make:221: *** target pattern contains no '%'. Stop. CMakeFiles/Makefile2:109: recipe for target 'CMakeFiles/nvonnxparser.dir/all' failed make[1]: *** [CMakeFiles/nvonnxparser.dir/all] Error 2 Makefile:151: recipe for target 'all' failed make: *** [all] Error 2
I do not know how to solve this issue. Can someone help me with this? Any type of help will be appreciated.
(Alternatively, for the setup part, do I need to download all 27 assets in the initial release link?)
Thank you.
The third dimension in
LiveChess2FEN/lc2fen/infer_pieces.py
Line 80 in 7693a47
__PREDS_DICT
. This could be made clearer.I wanted to try it out on my Jetson Nano, which has TRT 7.1. So I have to create the TRT model instead of using the one in Releases. When I run onnx2trt to convert:
./onnx2trt SqueezeNet1p1.onnx -o SqueezeNet1p1.trt
It asks me to provide an optimization profile:
----------------------------------------------------------------
Input filename: SqueezeNet1p1.onnx
ONNX IR version: 0.0.5
Opset version: 10
Producer name: keras2onnx
Producer version: 1.6.5
Domain: onnx
Model version: 0
Doc string:
----------------------------------------------------------------
Parsing model
Building TensorRT engine, FP16 available:1
Max batch size: 32
Max workspace size: 1024 MiB
[2021-03-22 22:50:16 ERROR] Network has dynamic or shape inputs, but no optimization profile has been defined.
[2021-03-22 22:50:16 ERROR] Network validation failed.
terminate called after throwing an instance of 'std::runtime_error'
what(): Failed to create object
Aborted (core dumped)
Could you please help me with this step?
After #39 is merged, add pytest setup for Ubuntu and basic testing instructions in the README.
Given the complexity and possible combinations in the code in lc2fen/infer_pieces.py
, it would be extremely beneficial to add some basic testing for (some of) the functions in that file.
To that end, a new test
directory could be created in the root of the repository to contain the test scripts. Two good python testing options could be unittest or pytest.
Feel free to leave a comment if you'd like some help/guidance tackling this problem or if you have some additional ideas.
Since the software stack supported by the Jetson Nano is no longer going to be updated, move all of the dependencies and code to a new branch to not hinder the development of new features of the project.
requirements_pc.txt
-> requirements.txt
Allow users to use miniconda to set up the Python environment and packages.
Per #34, we're going to have two versions, one for Jetson Nano and the other for a "standard desktop," but since the current version in the repo is for the standard desktop, should I go ahead and update the Python version to 3.11 and the TensorFlow version to 2.13.0?
(What about the other version? Should I write "version-agnostic" code so that the code works with both TensorFlow 2.13.0 and TensorFlow 2.5.0? Or do I not worry about the Jetson Nano's version for now?)
Creating a diagram of the project and how the different modules interact with each other would ease the development process. This could be added to the documentation to better navigate the project.
Create a Docker image to simplify setting up the development environment significantly. This can be done by following the setting up documentation for PC (Linux/Windows).
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.