#!

/usr/bin/python
#Copyright 2015 CVC-UAB

#This program is free software: you can redistribute it and/or modify

#it under the terms of the GNU General Public License as published by
#the Free Software Foundation, either version 3 of the License, or
#(at your option) any later version.

#This program is distributed in the hope that it will be useful,

#but WITHOUT ANY WARRANTY; without even the implied warranty of
#MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
#GNU General Public License for more details.

#You should have received a copy of the GNU General Public License
#along with this program. If not, see <http://www.gnu.org/licenses/>.

__author__ = "Miquel Ferrarons, David Vazquez"

__copyright__ = "Copyright 2015, CVC-UAB"
__credits__ = ["Miquel Ferrarons", "David Vazquez"]
__license__ = "GPL"
__version__ = "1.0"
__maintainer__ = "Miquel Ferrarons"

import numpy as np
import Config as cfg
import pickle
from sklearn import svm
from sklearn import linear_model

import os

def loadImageFeatures(featuresPath):
print 'Loading features from '+str(featuresPath)
file = open(featuresPath, 'r')
return pickle.load(file)

def run():
#List all the files .feat in the positives directory
positiveList = os.listdir(cfg.positiveFeaturesPath)
positiveList = filter(lambda element: '.feat' in element, positiveList)

#List all the files .feat in the negatives directory

negativeList = os.listdir(cfg.negativeFeaturesPath)
negativeList = filter(lambda element: '.feat' in element, negativeList)

#Count how many samples we have

positiveSamplesCount = len(positiveList)
negativeSamplesCount = len(negativeList)
samplesCount = positiveSamplesCount + negativeSamplesCount

#Load the features of the first element, to obtain the size of the feature
vectors.
filepath = cfg.positiveFeaturesPath + '/'+positiveList[0]
file = open(filepath, 'r')
feats = pickle.load(file)
featuresLength = len(feats)

#Initialize the structure that we will pass to the model for training
# X will be the samples for training
 # y will be the labels
X = np.zeros(shape=(samplesCount, featuresLength))
y = np.append(np.ones(shape=(1, positiveSamplesCount)),
-1*np.ones(shape=(1, negativeSamplesCount)))

# Load all the positive feature vectors in X

count = 0
for filename in positiveList:
filepath = cfg.positiveFeaturesPath+'/'+filename
X[count] = loadImageFeatures(filepath)
count += 1

# Load all the negative feature vectors in y

for filename in negativeList:
filepath = cfg.negativeFeaturesPath+'/'+filename
X[count] = loadImageFeatures(filepath)
count += 1

# Train the classifier with X and y, and some given parameters

if cfg.model == 'SVM':
print 'Training SVM....'
model = svm.LinearSVC(C=cfg.svm_C,
#loss='hinge',# loss
penalty=cfg.svm_penalty,
dual=cfg.svm_dual,
tol=cfg.svm_tol,
fit_intercept=cfg.svm_fit_intercept,
intercept_scaling=cfg.svm_intercept_scaling)
elif cfg.model == 'LogisticRegression':
print 'Training linear model....'
model = linear_model.LogisticRegression(C=cfg.logReg_C,
penalty=cfg.logReg_penalty,
dual=cfg.logReg_dual,
tol=cfg.logReg_tol,
fit_intercept=cfg.logReg_fit_intercept,
intercept_scaling=cfg.logReg_intercept_scaling)
else:
print 'ERROR: Model can only be SVM or LogisticRegression'
exit(0)

model.fit(X, y)
#Obtain the model score for the training set
print 'MODEL score'
print model.score(X, y)

#Save the model

modelDirectory, modelFilename = os.path.split(cfg.modelPath)
if not os.path.exists(modelDirectory):
os.makedirs(modelDirectory)

outputModelFile = open(cfg.modelPath, 'wb')

pickle.dump(model, outputModelFile)

if __name__ == '__main__':
run()