https://github.com/microsoft/CASPR
CASPR is a deep learning framework applying transformer architecture to learn and predict from tabular data at scale.
https://github.com/microsoft/CASPR
attention-mechanism business deep-learning tabular-data transformer transformer-architecture transformer-encoder
Last synced: 6 months ago
JSON representation
CASPR is a deep learning framework applying transformer architecture to learn and predict from tabular data at scale.
- Host: GitHub
- URL: https://github.com/microsoft/CASPR
- Owner: microsoft
- License: mit
- Created: 2022-11-03T16:16:14.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2023-02-09T21:51:46.000Z (over 2 years ago)
- Last Synced: 2025-03-30T21:37:48.689Z (6 months ago)
- Topics: attention-mechanism, business, deep-learning, tabular-data, transformer, transformer-architecture, transformer-encoder
- Language: Python
- Homepage:
- Size: 2.47 MB
- Stars: 38
- Watchers: 7
- Forks: 3
- Open Issues: 3
-
Metadata Files:
- Readme: README.md
- Contributing: CONTRIBUTING.md
- License: LICENSE
- Code of conduct: CODE_OF_CONDUCT.md
- Citation: CITATION.cff
- Security: SECURITY.md
- Support: SUPPORT.md
Awesome Lists containing this project
README
**CASPR is a transformer-based framework for deep learning from sequential data in tabular format, most common in business applications.**
Tasks critical to enterprise profitability, such as customer churn prediction, fraudulent account detection or customer lifetime value estimation, are often tackled by models trained on features engineered from customer data in tabular format. Application-specific feature engineering however adds development, operationalization and maintenance costs over time. Recent advances in representation learning present an opportunity to simplify and generalize feature engineering across applications.
With **CASPR** we propose a novel approach to encode sequential data in tabular format (e.g., customer transactions, purchase history and other interactions) into a generic representation of a subject's (e.g., customer's) association with the business. We evaluate these embeddings as features to train multiple models spanning a variety of applications (see: [paper](https://arxiv.org/abs/2211.09174)). CASPR, Customer Activity Sequence-based Prediction and Representation, applies transformer architecture to encode activity sequences to improve model performance and avoid bespoke feature engineering across applications. Our experiments at scale validate CASPR for both small and large enterprise applications.
## Getting Started & Resources
* **CASPR: Customer Activity Sequence-based Prediction and Representation** (NeurIPS 2022, New Orleans: Tabular Representation Learning)
- [paper](https://arxiv.org/abs/2211.09174)
- [poster](https://github.com/microsoft/CASPR/docs/images/caspr-poster.png)
* **Build**
- pre-requisites: ```python==3.9, setuptools```
- building the wheel: ```python setup.py build bdist_wheel```
* **Installation**
```
(now)
pip install .\dist\AI.Models.CASPR-.whl[]
(future)
pip install AI.Models.CASPR[]
```
use any of below modifiers, to customize the installation for target system / usecase:
```
horovod - for distributed training and inference on Horovod
databricks - for distributed training and inference on Databricks
aml - for (distributed) training and inference on Azure ML
hdi - for execution on Azure HD Insights
xai - to enable explainability
test - for extended test execution
dev - for development purposes only
```
* **Examples**
(TODO: can we point to a well commented one of our examples w/ or w/o data?)
## Contributions and Feedback
We welcome contributions! Please see the [contribution guidelines](CONTRIBUTING.md).
For feature requests or bug reports please file a [GitHub Issue](https://github.com/Microsoft/CASPR/issues).
## Code of Conduct
This project has adopted the [Microsoft Open Source Code of Conduct](https://opensource.microsoft.com/codeofconduct/).
For more information see the [Code of Conduct FAQ](https://opensource.microsoft.com/codeofconduct/faq/)
or contact [opencode@microsoft.com](mailto:opencode@microsoft.com) with any additional questions or comments.
## License
This project is licensed under the [MIT License](LICENSE).
---