| # Recommender Systems with Sparse Data |
| |
| This directory has a set of examples of how to build various kinds of recommender systems |
| using MXNet. It also includes a set of tools for using sparse data. |
| |
| ## Examples |
| |
| The examples are driven by notebook files. |
| |
| * [Matrix Factorization part 1: linear and non-linear models](demo1-MF.ipynb) |
| * [Matrix Factorization part 2: overfitting and deep ResNet](demo1-MF2-fancy.ipynb) |
| * [Binary classification with negative sampling](demo2-binary.ipynb) |
| * [Deep Structured Semantic Model (DSSM) for content-based recommendations](demo3-dssm.ipynb) |
| |
| ## Re-usable code |
| |
| These examples use and demonstrate a number of layers and other tools that can be used outside of these examples. They are all available from the [`recotools`](recotools.py) package. |
| |
| ### Negative Sampling |
| |
| * `NegativeSamplingDataIter` |
| |
| ### Loss Layers |
| |
| * `CosineLoss` |
| * `CrossEntropyLoss` |
| |
| ### Sparse Data Projection layers |
| |
| * `SparseRandomProjection` |
| * `SparseBagOfWordProjection` |
| |
| ## Acknowledgements |
| |
| Thanks to [xlvector](https://github.com/xlvector/) for the first Matrix Factorization example |
| that provided the basis for these examples. |
| |
| [MovieLens](http://grouplens.org/datasets/movielens/) data from [GroupLens](http://grouplens.org/). |
| |
