Different results without standardizing?

so-wise / weddell_gyre_clusters

Unsupervised classification of Weddell Gyre profiles

MIT License

2 stars 1 forks source link

Closed DaniJonesOcean closed 2 years ago

DaniJonesOcean commented 2 years ago

I wonder if we would get more clusters if we dropped the standardisation step, such that the actual values matter, instead of just the PC values.

DaniJonesOcean commented 2 years ago

Yeah, probably, but let's save this for future work. Closing for now.