alibaba/x-deeplearning
Alibaba's deep learning framework optimized for high-dimension sparse data in advertising and recommendation scenarios.

XDL is a complete solution for deep learning optimized for high-dimension sparse data scenarios commonly found in advertising, recommendation, and search systems. It provides a training engine with storage and communication optimizations, a prediction serving engine called Blaze, and algorithmic solutions including TDM (Deep Tree Matching) for recommendation matching and recall. The framework supports streaming training with feature admission and eviction capabilities.
Frequently asked
- What is alibaba/x-deeplearning?
- Alibaba's deep learning framework optimized for high-dimension sparse data in advertising and recommendation scenarios.
- Is x-deeplearning open source?
- Yes — alibaba/x-deeplearning is open source, released under the Apache-2.0 license.
- What language is x-deeplearning written in?
- alibaba/x-deeplearning is primarily written in PureBasic.
- How popular is x-deeplearning?
- alibaba/x-deeplearning has 4.3k stars on GitHub.
- Where can I find x-deeplearning?
- alibaba/x-deeplearning is on GitHub at https://github.com/alibaba/x-deeplearning.