How to use "Dirichlet Process Gaussian Mixture Model" in Scikit-Learn? (n_components?)

My understanding of "Infinite Mixture Models with Dirichlet Processes as Prior Distributions for Number of Clusters" is that the number of clusters is determined by the convergence of the data to a certain number of clusters.

This R Implementation https://github.com/jacobian1980/ecostates determines the number of clusters in this way. Not sure if that affects this effect though R implementationusing the Gibbs sampler.

What confuses me is the n_componentsparameter. n_components: int, default 1 : Number of mixture components. If the number of components is determined by the data and the Dirichlet process, what is this parameter?

Ultimately, I am trying to get:

(1) Cluster assignment of each sample;

(2) a probability vector for each cluster; and

(3) Likelihood/log-likelihood of each sample.

It looks like (1) is the predictmethod and (3) is the scoremethod. However, the output of (1) depends entirely on n_componentsthe hyperparameters.

My apologies if this is a naive question, I'm pretty new to Bayesian programming and found there is Dirichlet Processsomething Scikit-learnI'd like to try.

Here is the documentation : http://scikit-learn.org/stable/modules/generated/sklearn.mixture.DPGMM.html#sklearn.mixture.DPGMM

Here is a usage example : http://scikit-learn.org/stable/auto_examples/mixture/plot_gmm.html

Here is my naive usage:

from sklearn.mixture import DPGMM
X = pd.read_table("Data/processed/data.tsv", sep="\t", index_col=0)
Mod_dpgmm = DPGMM(n_components=3)
Mod_dpgmm.fit(X)