h2o4gpu
h2o4gpu copied to clipboard
h2oai
H2Oai GPU Edition
* There're a few copies of train dataset on GPU(CSC, CSR, COO sparse matrices). At least RMSE can be computed directly from CSC or CSR. * https://github.com/h2oai/h2o4gpu/pull/729
* add support of double, float16 and mixed precision(original data in float32 computation in float16 and so on) * https://github.com/h2oai/h2o4gpu/pull/729
Improvement of https://github.com/h2oai/h2o4gpu/pull/729
### Description After using the Gensim and PySpark LDA implementations, this is definitely an algorithm which could offer massive gains if developed for the GPU. For reference, the SciKit-Learn implementation...
Add a devel docker image (first for GPU, then for CPU when CPU only builds work). Like the jenkins one but also fetching and building the repo automatically.
Kmeans||
This is still WIP, for early review. TODOs: 1. Make the algorithm work as expected. 2. Add proper tests. 3. Make the algorithm work with k-means. 4. Factor out operations....
Initial implementation of field-aware factorization machines. Based on these 2 whitepapers: * https://arxiv.org/pdf/1701.04099.pdf * https://www.csie.ntu.edu.tw/~cjlin/papers/ffm.pdf And the following repositories: * https://github.com/guestwalk/libffm (original impl) * https://github.com/alexeygrigorev/libffm-python (Python interfact for it) *...
- **OS platform, distribution and version (e.g. Linux Ubuntu 16.04)**: Ubuntu 16.04.5 LTS (GNU/Linux 4.4.0-62-generic x86_64) - **Installed from (source or binary)**: Conda - **Version**: 0.3.0.9999 - **Python version (optional)**:...
Please do not use issues to ask questions. For questions use either [StackOverflow](https://stackoverflow.com/questions/tagged/h2o4gpu) (`h2o4gpu` tag) or [Gitter](https://gitter.im/h2oai/h2o4gpu?utm_source=share-link&utm_medium=link&utm_campaign=share-link). GitHub issues are used for: * Bug reporting. * Feature/enhancement requests and discussion....
