rdkit.DataManip.Metric.rdMetricMatrixCalc module¶
Module containing the calculator for metric matrix calculation, e.g. similarity and distance matrices
- rdkit.DataManip.Metric.rdMetricMatrixCalc.GetEuclideanDistMat((AtomPairsParameters)descripMat) object : ¶
Compute the distance matrix from a descriptor matrix using the Euclidean distance metric
ARGUMENTS:
- descripMat - A python object of any one of the following types
- A numeric array of dimensions n by m where n is the number of items in the data set
and m is the number of descriptors
- A list of Numeric Vectors (or 1D arrays), each entry in the list corresponds
to descriptor vector for one item
- A list (or tuple) of lists (or tuples) of values, where the values can be extracted to
double.
- RETURNS:
A numeric one-dimensional array containing the lower triangle elements of the symmetric distance matrix
- C++ signature :
_object* GetEuclideanDistMat(boost::python::api::object)
- rdkit.DataManip.Metric.rdMetricMatrixCalc.GetTanimotoDistMat((AtomPairsParameters)bitVectList) object : ¶
Compute the distance matrix from a list of BitVects using the Tanimoto distance metric
ARGUMENTS:
- bitVectList - a list of bit vectors. Currently this works only for a list of explicit bit vectors,
needs to be expanded to support a list of SparseBitVects
- RETURNS:
A numeric 1 dimensional array containing the lower triangle elements of the symmetric distance matrix
- C++ signature :
_object* GetTanimotoDistMat(boost::python::api::object)
- rdkit.DataManip.Metric.rdMetricMatrixCalc.GetTanimotoSimMat((AtomPairsParameters)bitVectList) object : ¶
Compute the similarity matrix from a list of BitVects
ARGUMENTS:
- bitVectList - a list of bit vectors. Currently this works only for a list of explicit bit vectors,
needs to be expanded to support a list of SparseBitVects
- RETURNS:
A numeric 1 dimensional array containing the lower triangle elements of the symmetric similarity matrix
- C++ signature :
_object* GetTanimotoSimMat(boost::python::api::object)