Add some fixes for Giotto and Meringue. Delauney has issues mainly cuz of the identical z-axis.... an easy fix. But maybe it's best to fix it from the dataset end (in the future).
Both methods need to be tested further for other datasets. Will update if other changes are required.
Also added a TODO for the thalamus dataset..apparently the label.tsv is not using the same column as that where the n_cluster is calculated. This can create a problem for metric evaluation.
Add some fixes for Giotto and Meringue. Delauney has issues mainly cuz of the identical z-axis.... an easy fix. But maybe it's best to fix it from the dataset end (in the future).
Both methods need to be tested further for other datasets. Will update if other changes are required.
Also added a TODO for the thalamus dataset..apparently the
label.tsvis not using the same column as that where then_clusteris calculated. This can create a problem for metric evaluation.