#! /usr/bin/env python3
#
def wave_regression_ols ( ):

#*****************************************************************************80
#
## wave_regression_ols() uses ordinary least squares regression on wave data.
#
#  Licensing:
#
#    This code is distributed under the MIT license.
#
#  Modified:
#
#    17 June 2023
#
#  Author:
#
#    Andreas Mueller, Sarah Guido.
#    Modifications by John Burkardt.
#
#  Reference:
#
#    Andreas Mueller, Sarah Guido,
#    Introduction to Machine Learning with Python,
#    OReilly, 2017,
#    ISBN: 978-1-449-36941-5
#
  import matplotlib.pyplot as plt
  import mglearn
  import numpy as np
  import platform
  import sklearn

  print ( '' )
  print ( 'wave_regression_ols():' )
  print ( '  Python version: ' + platform.python_version ( ) )
  print ( '  scikit-learn version: '+ sklearn.__version__ )
#
#  Generate the dataset.
#
  print ( '' )
  print ( '  Generate the wave dataset, (X, y).' )
  X, y = mglearn.datasets.make_wave ( n_samples = 60 )
  print ( "  X.shape:", X.shape )
#
#  Plot the dataset.
#
  print ( '  Plot the dataset.' )

  plt.clf ( )
  plt.plot ( X, y, 'o' )
  plt.ylim ( -3.0, +3.0 )
  plt.grid ( True )
  plt.xlabel ( "Feature" )
  plt.ylabel ( "Target" )
  filename = "wave_regression_data.png"
  plt.savefig ( filename )
  print ( "  Graphics saved as '" + filename + "'" )
#
#  Split the dataset.
#
  from sklearn.model_selection import train_test_split
  X_train, X_test, y_train, y_test = train_test_split ( X, y, random_state = 42 )
#
#  Create the model.
#
#  Note that the model "lr" will include many components or members
#  that can be accessed using the "lr." prefix.  
#  In particular, the coefficients and intercept of the linear model
#  are then created by calling "lr.fit()", and are then accessible
#  as lr.coef_ and lr.intercept_.
#
#
  from sklearn.linear_model import LinearRegression
  lr = LinearRegression ( )
  lr.fit ( X_train, y_train )

  a = lr.coef_
  b = lr.intercept_
  print ( )
  print ( '  Produce linear regression coefficients y = a * x + b' )
  print ( '  a = ', a, '  b = ', b )
  print ( '' )
  print ( '  Training score:' )
  print ( lr.score ( X_train, y_train ) )
  print ( '  Test score:' )
  print ( lr.score ( X_test, y_test ) )
#
#  Compare data to least squares line.
#
  print ( '' )
  print ( 'Compare least squares line to data.' )
  plt.plot ( X, y, 'bo' )
  Xmin = -3.0
  Ymin = a * Xmin + b
  Xmax = +3.0
  Ymax = a * Xmax + b
  plt.plot ( [Xmin,Xmax], [Ymin,Ymax], 'r-' )
  plt.ylim ( Xmin, Xmax )
  plt.grid ( True )
  plt.xlabel ( "Feature" )
  plt.ylabel ( "Target" )
  filename = "wave_regression_ols.png"
  plt.savefig ( filename )
  print ( "  Graphics saved as '" + filename + "'" )
#
#  Terminate.
#
  print ( '' )
  print ( 'wave_regression_ols():' )
  print ( '  Normal end of execution.' )

  return

def timestamp ( ):

#*****************************************************************************80
#
## timestamp() prints the date as a timestamp.
#
#  Licensing:
#
#    This code is distributed under the MIT license. 
#
#  Modified:
#
#    21 August 2019
#
#  Author:
#
#    John Burkardt
#
  import time

  t = time.time ( )
  print ( time.ctime ( t ) )

  return

if ( __name__ == '__main__' ):
  timestamp ( )
  wave_regression_ols ( )
  timestamp ( )