In the long run, depending on the number of metrics, we should consider a visual 2D dimensionality reduction (t-SNE?) of the selected metrics to identify possible correlations. However, before we do this, we need to test on real data whether this adds any value at all.
Do you think this could add value to the plugin or would that possibly be an overload @jxchen01?
In the long run, depending on the number of metrics, we should consider a visual 2D dimensionality reduction (t-SNE?) of the selected metrics to identify possible correlations. However, before we do this, we need to test on real data whether this adds any value at all.
Do you think this could add value to the plugin or would that possibly be an overload @jxchen01?