Hi all,
Browsing the source code I found that the AgglomerativeClustering
class appears to perform most forms of hierarchical clustering (not
only Ward as before). Firstly, I would like to ask if the code is
already "stable". I know that it is unreleased, but since it is on the
repository I assume that it should be pretty close to the final
version which will be released.
Also, I'm still not 100% sure how to use it based on the
documentation. More specifically, I'm interested in the case of using
a precomputed distance matrix. Thus, I want to confirm if the correct
to use the class would be something like:
from sklearn.cluster import hierarchical
dataset = np.genfromtxt('somedatafile)
X = my_pairwise_distance(dataset)
agg_cluster = hierarchical.AgglomerativeClustering(affinity='precomputed',
linkage='complete')
agg_cluster.fit(X)
...
To simplify, I just need to know if I can pass the distance matrix to
the fit method. The documentation is unclear on this, but checking the
source code it appears that it should work, since the linkage_tree
method can deal with the distance matrix.
Explaining why the documentation is unclear, it simply says on the fit method:
"
X : array-like, shape = [n_samples, n_features]
The samples a.k.a. observations.
"
It does not state that X can be a distance matrix in the case of precomputed.
Thanks in advance,
Flavio
------------------------------------------------------------------------------
"Accelerate Dev Cycles with Automated Cross-Browser Testing - For FREE
Instantly run your Selenium tests across 300+ browser/OS combos.
Get unparalleled scalability from the best Selenium testing platform available
Simple to use. Nothing to install. Get started now for free."
http://p.sf.net/sfu/SauceLabs
_______________________________________________
Scikit-learn-general mailing list
Scikit-learn-general@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/scikit-learn-general
