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

#*****************************************************************************80
#
## ram_regression_decision() uses a decision tree for RAM price regression.
#
#  Licensing:
#
#    This code is distributed under the MIT license.
#
#  Modified:
#
#    19 July 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
#
  from sklearn.tree import export_graphviz
  import graphviz
  import matplotlib.pyplot as plt
  import mglearn
  import numpy as np
  import os
  import pandas as pd
  import platform
  import sklearn

  print ( '' )
  print ( 'ram_regression_decision():' )
  print ( '  Python version: ' + platform.python_version ( ) )
  print ( '  scikit-learn version: '+ sklearn.__version__ )
#
#  Read the dataset.
#
  print ( '' )
  print ( '  Retrieve the RAM price dataset, (X, y).' )

  ram_prices = pd.read_csv ( 'ram_price.csv' )
#
#  Plot the data.
#
  plt.clf ( )
  plt.semilogy ( ram_prices.date, ram_prices.price )
  plt.grid ( True )
  plt.xlabel ( 'Year' )
  plt.ylabel ( 'Price in $/Mbyte' )
  plt.title ( 'Price for RAM Megabyte, semilog plot.' )
  filename = 'ram_data.png'
  plt.savefig ( filename )
  print ( '  Graphics saved as "' + filename + '"' )

  data_train = ram_prices[ram_prices.date < 2000]
  data_test = ram_prices[ram_prices.date >= 2000]
#
#  Predict prices.
#
  X_train = np.array ( data_train.date )
  X_train = X_train[:,np.newaxis]
  y_train = np.log ( data_train.price )

  from sklearn.tree import DecisionTreeRegressor

  tree = DecisionTreeRegressor ( max_depth = 3 ).fit ( X_train, y_train )

  X_all = np.array ( ram_prices.date )
  X_all = X_all[:,np.newaxis]

  pred = tree.predict ( X_all )
  price_tree = np.exp ( pred )

  plt.clf ( )
  plt.semilogy ( data_train.date, data_train.price, label = 'Training data' )
  plt.semilogy ( data_test.date, data_test.price, label = 'Test data' )
  plt.semilogy ( ram_prices.date, price_tree, label = 'Decision tree prediction' )
  plt.grid ( True )
  plt.xlabel ( 'Year' )
  plt.ylabel ( 'Price in $/Mbyte' )
  plt.title ( 'Price for RAM Megabyte, semilog plot.' )
  plt.legend ( )
  filename = 'ram_prediction.png'
  plt.savefig ( filename )
  print ( '  Graphics saved as "' + filename + '"' )
#
#  Terminate.
#
  print ( '' )
  print ( 'ram_regression_decision():' )
  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 ( )
  ram_regression_decision ( )
  timestamp ( )