Github user dusenberrymw commented on a diff in the pull request:
https://github.com/apache/spark/pull/7554#discussion_r35244705
--- Diff: python/pyspark/mllib/linalg.py ---
@@ -1152,9 +1156,385 @@ def sparse(numRows, numCols, colPtrs, rowIndices,
values):
return SparseMatrix(numRows, numCols, colPtrs, rowIndices, values)
+class DistributedMatrix(object):
+ """Represents a distributively stored matrix backed by one or more
RDDs."""
+ def numRows(self):
+ """Get or compute the number of rows."""
+ raise NotImplementedError
+
+ def numCols(self):
+ """Get or compute the number of cols."""
+ raise NotImplementedError
+
+
+class DistributedMatrices(object):
+ """Factory methods for distributed matrices."""
+ @staticmethod
+ def rowMatrix(rows, numRows=0, numCols=0):
+ """
+ Create a RowMatrix.
+
+ :param rows: An RDD of Vectors.
+ """
+ javaRowMatrix = callMLlibFunc("createRowMatrix", rows,
long(numRows), int(numCols))
+ jrm = JavaModelWrapper(javaRowMatrix)
+ return RowMatrix(jrm)
+
+ @staticmethod
+ def indexedRowMatrix(rows, numRows=0, numCols=0):
+ """
+ Create an IndexedRowMatrix.
+
+ :param rows: An RDD of IndexedRows or (long, Vector) tuples.
+ """
+ # We use DataFrames for serialization of IndexedRows from Python,
so convert the RDD to a
+ # DataFrame. This will convert each IndexedRow to a Row containing
the 'index' and 'vector'
+ # values, which can both be easily serialized. We will convert
back to IndexedRows on the
+ # Scala side.
+ javaIndexedRowMatrix = callMLlibFunc("createIndexedRowMatrix",
rows.toDF(),
+ long(numRows), int(numCols))
+ jirm = JavaModelWrapper(javaIndexedRowMatrix)
+ return IndexedRowMatrix(jirm)
+
+ @staticmethod
+ def coordinateMatrix(entries, numRows=0, numCols=0):
+ """
+ Create a CoordinateMatrix.
+
+ :param entries: An RDD of MatrixEntry inputs or (long, long,
float) tuples.
+ """
+ # We use DataFrames for serialization of MatrixEntry inputs from
Python, so convert the RDD
+ # to a DataFrame. This will convert each MatrixEntry to a Row
containing the 'i', 'j', and
+ # 'value' values, which can each be easily serialized. We will
convert back to MatrixEntry
+ # inputs on the Scala side.
+ javaCoordinateMatrix = callMLlibFunc("createCoordinateMatrix",
entries.toDF(),
+ long(numRows), long(numCols))
+ jcm = JavaModelWrapper(javaCoordinateMatrix)
+ return CoordinateMatrix(jcm)
+
+
+class RowMatrix(DistributedMatrix):
+ """
+ Represents a row-oriented distributed Matrix with no meaningful row
indices.
+
+ .. note:: Experimental
+ """
+ def __init__(self, jrm):
+ """ Create a wrapper over a Java RowMatrix. """
+ self._jrm = jrm
+ self.rows = self._rows()
+
+ def _rows(self):
+ """
+ Get the rows of the RowMatrix as a RDD of Vectors.
+
+ >>> rows = sc.parallelize([Vectors.dense([1, 2, 3]),
Vectors.dense([4, 5, 6]),
+ ... Vectors.dense([7, 8, 9]),
Vectors.dense([10, 11, 12])])
+ >>> rm = DistributedMatrices.rowMatrix(rows)
+ >>> rowsRDD = rm.rows
+ >>> rowsRDD.first()
+ DenseVector([1.0, 2.0, 3.0])
+ """
+ return self._jrm.call("rows")
+
+ def numRows(self):
+ """
+ Get or compute the number of rows.
+
+ >>> rows = sc.parallelize([Vectors.dense([1, 2, 3]),
Vectors.dense([4, 5, 6]),
+ ... Vectors.dense([7, 8, 9]),
Vectors.dense([10, 11, 12])])
+ >>> rm = DistributedMatrices.rowMatrix(rows)
+ >>> int(rm.numRows())
+ 4
+
+ >>> rows = sc.parallelize([Vectors.dense([1, 2, 3]),
Vectors.dense([4, 5, 6]),
+ ... Vectors.dense([7, 8, 9]),
Vectors.dense([10, 11, 12])])
+ >>> rm = DistributedMatrices.rowMatrix(rows, 7, 6)
+ >>> int(rm.numRows())
+ 7
+ """
+ return self._jrm.call("numRows")
+
+ def numCols(self):
+ """
+ Get or compute the number of cols.
+
+ >>> rows = sc.parallelize([Vectors.dense([1, 2, 3]),
Vectors.dense([4, 5, 6]),
+ ... Vectors.dense([7, 8, 9]),
Vectors.dense([10, 11, 12])])
+ >>> rm = DistributedMatrices.rowMatrix(rows)
+ >>> int(rm.numCols())
+ 3
+
+ >>> rows = sc.parallelize([Vectors.dense([1, 2, 3]),
Vectors.dense([4, 5, 6]),
+ ... Vectors.dense([7, 8, 9]),
Vectors.dense([10, 11, 12])])
+ >>> rm = DistributedMatrices.rowMatrix(rows, 7, 6)
+ >>> int(rm.numCols())
+ 6
+ """
+ return self._jrm.call("numCols")
+
+
+class IndexedRow(object):
+ """
+ Represents a row of an IndexedRowMatrix. Just a wrapper over a (long,
Vector) tuple.
+
+ .. note:: Experimental
+ """
+ def __init__(self, index, vector):
+ self.index = long(index)
+ self.vector = _convert_to_vector(vector)
+
+ def __repr__(self):
+ return "IndexedRow(%s, %s)" % (self.index, self.vector)
+
+
+class IndexedRowMatrix(DistributedMatrix):
+ """
+ Represents a row-oriented distributed Matrix with indexed rows.
+
+ .. note:: Experimental
+ """
+ def __init__(self, jirm):
+ """ Create a wrapper over a Java IndexedRowMatrix. """
+ self._jirm = jirm
+ self.rows = self._rows()
+
+ def _rows(self):
--- End diff --
Yeah, the `_rows()` method is private since it is just the implementation
for the variable `self.rows`, which mimics the `val rows` in Scala. Pulling
the logic out just makes the code cleaner, which is more important for
`IndexedRowMatrix` and `CoordinateMatrix` than for `RowMatrix`.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
