Code source de DummyClassifier

#!/usr/bin/env python
# -*- encoding: utf-8 -*-
""" A dummy model, with 3 strategies :
 
 * stratified;
 * most_frequent;
 * uniform.

The doc is here : http://scikit-learn.org/dev/modules/generated/sklearn.dummy.DummyClassifier.html#sklearn.dummy.DummyClassifier

--------------------------------------------------------------------------------

Sortie du script
----------------
.. runblock:: console

    $ python DummyClassifier.py

Résultats
---------
La soumission du résultat à Kaggle donne ??.??% (pas encore fait).

--------------------------------------------------------------------------------
"""
__author__	= 'Lilian BESSON (mailto:lilian.besson[AT]normale.fr)'

from KaggleModel import *

################################################################################
# Beginning to learn

from sklearn.dummy import DummyClassifier
from sklearn.utils import shuffle

################################################################################
# ok, let use this 'cross validation' process to find the best
# meta parameter : strategy
strategy_quality = {}
list_strategy = ['stratified', 'most_frequent', 'uniform']	#: Espace de recherche
Number_try = 100		#: Nombre de tests utilisés pour méta-apprendre
proportion_train = 0.75		#: Proportion d'individus utilisés pour méta-apprendre.
print("Find the best value for the meta parameter strategy, with %i run for each..." % Number_try)
print("Searching in the range : %s..." % str(list_strategy))

print("""Using the first part (%2.2f%%, %i passengers) of the training dataset as training, 
and the second part (%2.2f%%, %i passengers) as testing !"""
 % ( 100.0*proportion_train, int(number_passengers*proportion_train),
     100.0*(1-proportion_train), number_passengers - int(number_passengers*proportion_train) ))

for strategy in list_strategy:
#	train_data = shuffle(train_data)
	Dummy = DummyClassifier(strategy = strategy)
	print("For the strategy %s, learning from the first part of the dataset..." % strategy)
	quality=[]
	for nb_essais in xrange(Number_try):
#		train_data = shuffle(train_data)
		Dummy = Dummy.fit(train_data[0:int(number_passengers*proportion_train),1::],
		 train_data[0:int(number_passengers*proportion_train),0])
		Output = Dummy.predict(train_data[number_passengers - int(number_passengers*proportion_train)::,1::])
		quality.append(100.0 * Output[Output == train_data[number_passengers - int(number_passengers*proportion_train)::,0]].size / Output.size)
	strategy_quality[strategy] = np.mean(quality)
	print("... this value of strategy seems to have a (mean) quality = %2.2f%%..." % np.mean(quality))

val = strategy_quality.values()
#: La valeur optimale trouvée pour le paramètre strategy
best_strategy = strategy_quality.keys()[val.index(np.max(val))]
print("With trying each of the following strategy (%s), each %i times, the best one is %s. (for a quality = %2.2f%%)"
 % (str(list_strategy), Number_try, best_strategy, np.max(val)))


################################################################################
# And then use this 'best' value of strategy to predict on the test dataset
print("Creating the Dummy Classifier classifier with best meta parameters.")
Dummy = DummyClassifier(strategy = best_strategy)
print("Learning...")
Dummy = Dummy.fit(train_data[0::,1::],train_data[0::,0])
#: The score for this classifier.
score = (100.0*Dummy.score(train_data[0::,1::], train_data[0::,0]) )
print(" Proportion of perfect fitting for the training dataset = %2.2f%%" % 
 score )
# ~ must be < 95%

# Predict on the testing set
test_file_object = csv.reader(open('test.csv', 'rb'))
header = test_file_object.next()

print("Predicting for the testing dataset")
Output = Dummy.predict(test_data) 

# Write the output
open_file_object = csv.writer(open("csv/Dummy_best.csv", "wb"))

z = 0
for row in test_file_object:
 row.insert(0, int(Output[z])) # Insert the prediction at the start of the row
 open_file_object.writerow(row) # Write the row to the file
 z += 1

print("Prediction: wrote in the file csv/Dummy_best.csv.")