pydbm.clustering.computableclusteringloss.repellingloss package


pydbm.clustering.computableclusteringloss.repellingloss.improved_repelling_loss module

class pydbm.clustering.computableclusteringloss.repellingloss.improved_repelling_loss.ImprovedRepellingLoss

Bases: pydbm.clustering.computableclusteringloss.repelling_loss.RepellingLoss

Repelling Loss.

Note that this class calculates this penalty term for each cluster divided by soft assignments and refers to the sum as a regularizer.


  • Aljalbout, E., Golkov, V., Siddiqui, Y., Strobel, M., & Cremers, D. (2018). Clustering with deep learning: Taxonomy and new methods. arXiv preprint arXiv:1801.07648.
  • Guo, X., Gao, L., Liu, X., & Yin, J. (2017, June). Improved Deep Embedded Clustering with Local Structure Preservation. In IJCAI (pp. 1753-1759).
  • Xie, J., Girshick, R., & Farhadi, A. (2016, June). Unsupervised deep embedding for clustering analysis. In International conference on machine learning (pp. 478-487).
  • Zhao, J., Mathieu, M., & LeCun, Y. (2016). Energy-based generative adversarial network. arXiv preprint arXiv:1609.03126.

Compute clustering loss.

  • observed_arrnp.ndarray of observed data points.
  • reconstructed_arrnp.ndarray of reconstructed data.
  • feature_arrnp.ndarray of feature points.
  • delta_arrnp.ndarray of differences between feature points and centroids.
  • p_arrnp.ndarray of result of soft assignment.
  • q_arrnp.ndarray of target distribution.

Tuple data. - np.ndarray of delta for the encoder. - np.ndarray of delta for the decoder. - np.ndarray of delta for the centroids.

Module contents