I'm opening this issue to discuss the possible inclusion of Geometric-SMOTE, proposed by Douzas and Bacao [in this paper](https://www.sciencedirect.com/science/article/pii/S0020025519305353), in the imbalanced-learn library. The oversampler was already implemented by Georgios Douzas [in this repository](https://github.com/georgedouzas/geometric-smote/). It is compatible with the scikit/imbalanced-learn libraries and contains a test suite similar to the ones that already exist for SMOTE-based oversamplers. In addition, his implementation has a MIT license. Considering that this oversampler is essentially a generalization of the generation mechanism of SMOTE (in fact, given specific hyperparameters, it mimics the behavior of SMOTE) that appears to have a consistent performance, I believe it would be a nice addition to this library. I recently discussed this idea with both authors, which also approved the idea. #### Describe the solution you'd like Inclusion of the Geometric-SMOTE oversampler in the imbalanced-learn library. I would be happy to do this. I will make a PR referencing this issue soon. Please let me know if there is any additional information I should consider before proceeding.