Type Error

[kangotti]

hello,

I am trying to run this algorithm on my dataset but I am getting this TypeError.

TypeError Traceback (most recent call last) in 22 intersection = umap.umap_.general_simplicial_set_intersection(fit1.graph_, fit2.graph_, weight=categorical_weight) 23 intersection = umap.umap_.reset_local_connectivity(intersection) ---> 24 embedding = umap.umap_.simplicial_set_embedding(fit1._raw_data, intersection, fit1.n_components, 25 fit1._initial_alpha, fit1._a, fit1._b, 26 fit1.repulsion_strength, fit1.negative_sample_rate,

TypeError: simplicial_set_embedding() missing 2 required positional arguments: 'densmap_kwds' and 'output_dens'

[lmcinnes]

The newer code has some extra parameters to that function (which is largely an internal function). On the plus side you should now be able to do something like

combined = fit1 * fit2

to get the same result as what you seem to be hand coding there.

[arnaud-nt2i]

@lmcinnes Could you elaborate on how to how to use new combined fits?

To everyone else reading this: the two new parameters should be filled by respectively "{}" and "False"

[lmcinnes]

I believe you should be able to see more details on the combined fits in the documentation: https://umap-learn.readthedocs.io/en/latest/composing_models.html

[arnaud-nt2i]

@lmcinnes thank you. Sorry for RTFM issu...

[gaurav-95]

Also as a consequence when plotting the scatter map the result has to be converted to an array and the first index should be plotted:

Ex: plt.scatter(*np.array(embedding)[0].T, s=2, cmap='Spectral', alpha=1.0)

[epkuznetsova]

I passed the following arguments. Might not be the right thing to do, but it got the job done: (..., densmap_kwds = {}, output_dens = False)