Transforming new data points puts them on the outskirts of existing clusters ...

Nobody has claimed this yet.

... but never "inside" of clusters.

The blue points in the figure are "new" points that are transformed into the existing projection of the points of other colors.


Is there something I am doing wrong?

```python
config = {
'random_state': 42,
'n_neighbors': 10,
'n_components': 2,
'metric': 'cosine',
'spread': 2,
'min_dist': 0.01
}

trans = umap.UMAP(**config)
trans = trans.fit(a) # fit on some points
projection_a = trans.transform(a) # project them
projection_b = trans.transform(b) # project the "new" points
```

Thank you for your help!

Start by reproducing the issue from the provided `umap.UMAP` configuration and the `fit`, `transform(a)`, and `transform(b)` calls. Inspect the implementation and documentation for `transform` to determine whether the placement of new points is expected or indicates a defect; done means the behavior is explained or a reproducible failing test and scoped fix are identified.

Written by the indexing model from the issue text.