Buffalo

- Added new algorithm: `Probabilistic Latent Semantic Indexing(pLSI)`
- Support warm-start

- Delete existing temporary data file when creating new data file

- Removed tensorflow dependency

- Remove python n2 dependency
- Still needs in testing
- Resolved build time dependency error

- Calculate idcg indpendently of rec result

- Fix bug of W2V
- OOV caused by `min_cut`

Updated
- Improved the speed of database creation.

Bug Fixes
- Fixed warp accuracy test code
- Fix typos

Update
- Added new algorithm: Weighted Approximate-Rank Pairwise Matrix Factorization
- Weston, Jason, Samy Bengio, and Nicolas Usunier. "Wsabie: Scaling up to large vocabulary image annotation." Twenty-Second International Joint Conference on Artificial Intelligence. 2011.
- Check-out [some experiments](https://github.com/kakao/buffalo/blob/dev/benchmark/accuracy_warp.md), and [unit test code](https://github.com/kakao/buffalo/blob/master/tests/algo/test_warp.py)
- Added accuracy benchmarking results (work in progress)

Bug fixes
- fix ImportError for 3rd/n2

**WARNING: THIS IS BROKEN VERSION**

Bug Fixes
- Fix failure with creating a large database
- Fix bug that get_most_similar_item returns Nan values.

Misc
- Clean Code(Linting, Refactoring)
- Add new Jupiter example explaining [large scale recommendation with Buffalo](https://github.com/kakao/buffalo/blob/dev/examples/jupyter-examples/6.%20KakaoBrunch12M.ipynb)

Update
- Add CUDA accelerated [BPR-MF](https://github.com/kakao/buffalo/blob/e1dd708a17e265f3f64a4f094e1be8eef89ceee2/buffalo/algo/options.py#L230)
[sample usage](https://github.com/kakao/buffalo/blob/e1dd708a17e265f3f64a4f094e1be8eef89ceee2/tests/algo/test_bpr.py#L89-L118)
- When tested against ml-20m and Kakao Brunch Dataset, CUDA version of BPR-MF was about 5x times faster than CPU implementation of BPR-MF in training time.


- Add metric [truncated AUC](https://wiki.epfl.ch/edicpublic/documents/Candidacy%20exam/Evaluation.pdf)
- Add [notebook examples](https://github.com/kakao/buffalo/tree/v1.1.0/examples/jupyter-examples)
- Update documents



Bug fixes
- Fix SGD update of BPR
- Fix to use bias when calculating top-k recommendation in BPR-MF


Misc
- Clean code(linting, refactoring)
- Add internal data type check and state check
- Change validation format to be CSR matrix
- Validation took about 40% faster than the previous version.
- but this might be harmful because it requires larger memory during validation.