scripts/builtin/differenceStatistics.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.
#
#-------------------------------------------------------------

# Prints the difference statistics of two matrices given, to indicate how
# they are different. This can be used for instance in comparison of lossy
# compression techniques, that reduce the fidelity of the data. 
#
# INPUT:
# --------------------------------------------------------------------------------
# X        First Matrix to compare
# Y        Second Matrix to compare
# --------------------------------------------------------------------------------

m_differenceStatistics = function(Matrix[Double] X, Matrix[Double] Y)  {

  P = matrix("0.0 0.01 0.1 0.25 0.5 0.75 0.90 0.99 1.0", rows= 9, cols=1)
  print("quantiles are: " + toString(target=P, linesep=" "))
  
  # Mean Square Error
  [MSE, SEQ] = mse(X=X, Y=Y, P=P)
  print("                Mean Square Error: " + toString(as.scalar(MSE)))
  print("            Quantile Square Error: " + toString(target=SEQ , linesep=" "))

  # Root Mean Square Error
  [RMSE, RSEQ] = rmse(X=X, Y=Y, P=P)
  print("           Root Mean Square Error: " + toString(as.scalar(RMSE)))
  print("       Quantile Root Square Error: " + toString(target=RSEQ , linesep=" "))

  # Normalized Root Mean Square Error
  [NRMSE, NRSEQ] = nrmse(X=X, Y=Y, P=P)
  print("Normalized Root Mean Square Error: " + toString(as.scalar(NRMSE)))
  print("  Quantile Norm Root Square Error: " + toString(target=NRSEQ , linesep=" "))

  # Mean Absolute Error
  [MAE, AEQ] = mae(X=X, Y=Y, P=P)
  print("              Mean Absolute Error: " + toString(as.scalar(MAE)))
  print("          Quantile Absolute Error: " + toString(target=AEQ , linesep=" "))

  # Mean Absolute Percentage Error
  [MAPE, APEQ] = mape(X=X, Y=Y, P=P)
  print("   Mean Absolute percentage Error: " + toString(as.scalar(MAPE)))
  print("                     Quantile APE: " + toString(target=APEQ , linesep=" "))

  # Symmetric Mean Absolute Percentage Error 
  [sMAPE, sAPEQ] = smape(X=X, Y=Y, P=P)
  print("symmetric Mean Absolute per Error: " + toString(as.scalar(sMAPE)))
  print("          Quantile symmetric MAPE: " + toString(target=sAPEQ , linesep=" "))

  # Modified Symmetric Mean Absolute Percentage Error
  [msMAPE, msAPEQ] = msmape(X=X, Y=Y, P=P)
  print("          modified symmetric MAPE: " + toString(as.scalar(msMAPE)))
  print(" Quantile modified symmetric MAPE: " + toString(target=msAPEQ , linesep=" "))


  # Peak Signal to Noise Ratio
  PSNR = psnr(X=X, Y=Y)
  print("       Peak Signal to Noise Ratio: " + toString(target=PSNR , linesep=" "))


  # Standard deviation
  sd_Y = sd(Y)
  sd_X = sd(X)
  sd_diff = abs(sd_Y - sd_X)
  print("  Standard Deviation Values (X,Y): (" + toString(sd_X) + ", " + toString(sd_Y) + ")")
  print("    Standard Deviation Difference: " + toString(sd_diff))

  skew_x = skewness(X)
  skew_y = skewness(Y)
  skew_diff = abs(skew_y - skew_x)
  print("                Skew Values (X,Y): (" + toString(target=skew_x, linesep=" ") + ", " + toString(target=skew_y, linesep=" ") + ")")
  print("                  Skew Difference: " + toString(skew_diff))

}