| Clustering Algorithms |
| dataset name |
Composable algorithms make it possible to perform parameter sweeps over regular clustering algorithms. When you choose a composable algorithm, you'll return to this page, so you can build a complex chain of composable algorithms that is finalized by simple algorithm. Once you finalize it with a simple algorithm, you will then be presented with a page allowing you to set all the parameters for these algorithms.
| Composable Clustering Algorithms |
| dataset name |
Here you can start to create a clustering algorithm run. You can go directly to a non-composable algorithm if you want to run a clustering algorithm with a known set of specific paramters.
The composable algorithms are used to do various types of parameter sweeps over a non-composable clustering algorithm. For instance MultiRun or MCCV can search over a set of values for k, and will attempt to select the one the that produces the most likely model.