Implementation for ICML 2024 Paper "Online Cascade Learning for Efficient Inference over Streams".

Installation

To use this project, follow these steps:

1. Clone the repository: git clone https://github.com/flitternie/online_cascade_learning.git
2. Install the required dependencies: conda env create -f requirements.yml

Usage

1. Prepare your dataset using Huggingface Datasets library. Preprocessed data for reproducing paper results are available here.
2. Create <dataset_name>.py under ./data/, specifying your data preprocess and postprocess functions. Read Data Module Specification for details.
3. Create .yaml configuration file to customize your cascade and hyperparameters. We currently support Scikit-learn classification models and Huggingface transformers sequence classification models to compose a cascade. Most common LLMs, like OpenAI GPT, Anthropic Claude, Google Gemini, Mistral, etc., are supported! Read Configuration File Specification for details.
4. Run online cascade learning: python run.py --config <path_to_config>.

Contributing

Contributions are welcome! If you find any issues or have suggestions for improvements, please open an issue or submit a pull request.

Active Contributors: @Flitternie @Dingz9926

Citation

@inproceedings{nie2024online,
  title={Online Cascade Learning for Efficient Inference over Streams},
  author={Nie, Lunyiu and Ding, Zhimin and Hu, Erdong and Jermaine, Christopher and Chaudhuri, Swarat},
  booktitle={International Conference on Machine Learning},
  year={2024},
  organization={PMLR}
}