scripts/builtin/pca.dml

#-------------------------------------------------------------
#
# Licensed to the Apache Software Foundation (ASF) under one
# or more contributor license agreements.  See the NOTICE file
# distributed with this work for additional information
# regarding copyright ownership.  The ASF licenses this file
# to you under the Apache License, Version 2.0 (the
# "License"); you may not use this file except in compliance
# with the License.  You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing,
# software distributed under the License is distributed on an
# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
# KIND, either express or implied.  See the License for the
# specific language governing permissions and limitations
# under the License.
#
#-------------------------------------------------------------

# Principal Component Analysis (PCA) for dimensionality reduction
# ---------------------------------------------------------------------------------------------
# NAME   TYPE    DEFAULT  MEANING
# ---------------------------------------------------------------------------------------------
# X      Matrix  ---      Input feature matrix
# K      Int     ---      Number of reduced dimensions (i.e., columns)
# Center Boolean TRUE     Indicates whether or not to center the feature matrix
# Scale  Boolean TRUE     Indicates whether or not to scale the feature matrix
# ---------------------------------------------------------------------------------------------
# Xout   Matrix  ---      Output feature matrix with K columns
# Mout   Matrix  ---      Output dominant eigen vectors (can be used for projections)
# ---------------------------------------------------------------------------------------------

m_pca = function(Matrix[Double] X, Integer K=2, Boolean center=TRUE, Boolean scale=TRUE)
  return (Matrix[Double] Xout, Matrix[Double] Mout) 
{
  N = nrow(X);
  D = ncol(X);

  # perform z-scoring (centering and scaling)
  X = scale(X, center, scale);

  # co-variance matrix
  mu = colSums(X)/N;
  C = (t(X) %*% X)/(N-1) - (N/(N-1))*t(mu) %*% mu;

  # compute eigen vectors and values
  [evalues, evectors] = eigen(C);

  decreasing_Idx = order(target=evalues,by=1,decreasing=TRUE,index.return=TRUE);
  diagmat = table(seq(1,D),decreasing_Idx);
  # sorts eigenvalues by decreasing order
  evalues = diagmat %*% evalues;
  # sorts eigenvectors column-wise in the order of decreasing eigenvalues
  evectors = evectors %*% diagmat;

  eval_dominant = evalues[1:K, 1];
  evec_dominant = evectors[,1:K];

  # Construct new data set by treating computed dominant eigenvectors as the basis vectors
  Xout = X %*% evec_dominant;
  Mout = evec_dominant;
}