iyerr3 commented on a change in pull request #352: Feature/kd tree knn
URL: https://github.com/apache/madlib/pull/352#discussion_r256158067
##########
File path: src/ports/postgres/modules/knn/knn.py_in
##########
@@ -169,29 +543,50 @@ def knn(schema_madlib, point_source, point_column_name,
point_id,
DOUBLE PRECISION[] x, DOUBLE
PRECISION[] y -> DOUBLE PRECISION
@param weighted_avg Calculates the Regression or
classication of k-NN using
the weighted average method.
+ @param algorithm The algorithm to use for knn
+ @param algorithm_params The parameters for kd-tree algorithm
"""
with MinWarning('warning'):
+
output_neighbors = True if output_neighbors is None else
output_neighbors
if k is None:
k = 1
+
+ algorithm = _get_algorithm_name(algorithm)
+
+ # Default values for depth and leaf nodes
+ depth = 3
+ leaf_nodes = 2
Review comment:
Could we rename this to something along the lines of
`max_leaves_to_explore`? I confused this with the actual number of leaf nodes
in the tree during first read.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services