EOFError by loading umap model

Nobody has claimed this yet.

I have created a umap model and saved it via joblib. Then I try to load the model on another PC but with the same environment. My memory is 50% full. Is it not possible to save an umap model with joblib?

```
umap_model_plot = umap.UMAP(n_neighbors=15,
n_components=2,
metric='cosine').fit(emb)

joblib.dump(umap_model_plot,'umap_2.sav')
```

Load model

```

joblib.load('umap_2.sav')
Traceback (most recent call last):

File "", line 1, in
joblib.load('umap_2.sav')

File "C:\Users\Paul\AppData\Roaming\Python\Python37\site-packages\joblib\numpy_pickle.py", line 585, in load
obj = _unpickle(fobj, filename, mmap_mode)

File "C:\Users\Paul\AppData\Roaming\Python\Python37\site-packages\joblib\numpy_pickle.py", line 504, in _unpickle
obj = unpickler.load()

File "C:\Users\Paul\Anaconda3\envs\top2vec_final\lib\pickle.py", line 1086, in load
raise EOFError

EOFError
```

Start by reproducing the shown umap.UMAP, joblib.dump, and joblib.load sequence in the reported Python 3.7 environments, then inspect how the saved file is created and read. Done means identifying a reproducible cause for the EOFError and documenting or validating a supported loading path.

Written by the indexing model from the issue text.