Open-source Platform for Engineering Neural Architectures and Research Collaboration. Developing and improving AI tools for everyone.
OTHER License
π Welcome to OPEN-ARC, an open-source initiative to further AI research through collaboration. This repository contains base model files and folders for various problem sets, each accompanied by a challenge and linked Kaggle notebooks to run the models. Users can link their notebooks to the challenges, share their findings, and help improve AI models to help solve problems in various fields and further research enhancement.
OPEN-ARC aims to democratize AI research by providing a collaborative platform where users can:
To start with OPEN-ARC, clone the repository or start from your notebook locally or in Kaggle or Colab using the provided datasets.
git clone https://github.com/infinitode/open-arc.git
cd open-arc
[!TIP] You can also download the available Jupyter notebooks and base models. They contain everything you need for that project, if you want to work on specific projects only.
Each project folder contains our:
[!NOTE] Base models are not available for all of the projects in OPEN-ARC.
To run a model, you can just navigate to the project folder and follow the instructions in the README.md file located there. You can also run the Kaggle notebook, either locally or in Kaggle/Colab (note that we will primarily use Kaggle Datasets, which means that you'd have to obtain the datasets either from Kaggle, or another source, to run the code locally or in Colab), and follow the steps inside.
We only provide the base models, and the basic code implementations, this is where the power of community comes in. You can either improve the base code or write your own. Then, others can learn from your implementation, therefore furthering research, and helping communities worldwide.
We welcome contributions from the community. To contribute to the project:
[!TIP] You can also work on a project and share your results by submitting a link to your notebook. If you'd like to have your entry listed on the leaderboard, fork the repo and update
LEADERBOARD.md
in your pull request.
To protect our community, please ensure your contributions adhere to our Code of Conduct.
Here are some of the current projects available in OPEN-ARC:
Rank | Contributor | Architecture Type | Platform | Base Model | Dataset | Accuracy | Link |
---|---|---|---|---|---|---|---|
N | Our Model | RandomForestClassifier | Kaggle | β | Liver Cirrhosis Stage Classification π©Ί | 95.6% | Notebook |
Rank | Contributor | Architecture Type | Platform | Base Model | Dataset | Accuracy | Link |
---|---|---|---|---|---|---|---|
N | Our Model | RandomForestClassifier | Kaggle | β | Weather Type Classification | 91.2% | Notebook |
Rank | Contributor | Architecture Type | Platform | Base Model | Dataset | Accuracy | Link |
---|---|---|---|---|---|---|---|
N | Our Model | CustomCNN | Kaggle | β | π± Potato Plant Diseases Data π | 95.1% | Notebook |
Rank | Contributor | Architecture Type | Platform | Base Model | Dataset | Accuracy | Link |
---|---|---|---|---|---|---|---|
N | Our Model | GradientBoostingClassifier | Kaggle | β | Red Wine Quality | 72.8% | Notebook |
Rank | Contributor | Architecture Type | Platform | Base Model | Dataset | Accuracy | Link |
---|---|---|---|---|---|---|---|
N | Our Model | SimpleRNN | Kaggle | β | All Terraria Weapons DPS V_1.4.4.9 | 78.6% | Notebook |
Rank | Contributor | Architecture Type | Platform | Base Model | Dataset | BLEU-Score | Link |
---|---|---|---|---|---|---|---|
N | Our Model | DistilBART | Kaggle | β | NEWS SUMMARY | 52.8% | Notebook |
Rank | Contributor | Architecture Type | Platform | Base Model | Dataset | Accuracy | Link |
---|---|---|---|---|---|---|---|
N | Our Model | XGBClassifier | Kaggle | β | Crop Recommendation Dataset | 98.6% | Notebook |
More projects will be added soon!
We encourage users to share their progress and improvements. You can do this in several ways:
Local Notebooks:
Google Colab:
Kaggle Notebooks:
Notebook Documentation:
Feel free to document your process and findings at the bottom of the project README for others to learn from and improve upon.
You can also upload your trained models and code to GitHub and share the links in our repository's leaderboard. Hereβs how:
Upload Your Model (optional):
Share the Link:
LEADERBOARD.md
file in the main repository.LEADERBOARD.md
file.Post on the Leaderboard:
Users can quickly share their contributions and help others in the community learn and improve their models and research.
This project is licensed under the MIT License - see the LICENSE file for more details about this repo's license.
We hope OPEN-ARC becomes a thriving community of developers, helping improve AI tools for communities around the world, and drive new research and technology.
Happy coding and collaborating!
~ Infinitode