def writePathToSamples(parentDir):
'''write paths to samples for each labels. The .txt files are generated
for each block and it is stored at
projectPath/baseNameOfparentDir/block_<number>.txt
Each .txt file contains paths to the samples of that block for
all the users'''
userlist = utility.getSubdirectories(parentDir)
print userlist
baseName = os.path.basename(parentDir)
data_dir = os.path.join(projectPath,baseName)
utility.checkDirectory(data_dir)
# blocks = ["{:02d}".format(x) for x in range(1,10)]
for i in range(1,10):
with open(os.path.join(data_dir,'block_'+str(i)+'.txt'),'w') as f:
for user in userlist:
blockPath = os.path.join(parentDir,user,'block_'+str(i))
if os.path.isdir(blockPath):
img_list = os.listdir(blockPath)
for img in img_list:
img_path = os.path.join(blockPath,img)
if not os.path.isfile(img_path) or (img_path.split('.')[-1] != "jpg"
and img_path.split('.')[-1] != "png"):
continue
f.write("%s\n"%img_path)
def classify_svm(feature, labels, model='left'):
print "---------SVM Classifier-------------"
modelDir = os.path.join(projectPath, model)
utility.checkDirectory(modelDir)
dataTrain, dataTest, labelsTrain, labelsTest = train_test_split(
feature, labels, test_size=0.20, random_state=42)
param_grid = [
{
'C': [1, 10, 100, 1000],
'gamma': [1, 0.1, 0.001, 0.0001],
'kernel': ['linear']
},
{
'C': [1, 10, 100, 1000],
'gamma': [1, 0.1, 0.001, 0.0001],
'kernel': ['rbf']
},
]
svc = svm.SVC()
clf = grid_search.GridSearchCV(estimator=svc,
param_grid=param_grid,
cv=5,
n_jobs=-2)
print "Training SVM classifier for grid of C and gamma values to select best parameter\n"
start = time.time()
clf.fit(dataTrain, labelsTrain)
end = time.time()
elapsed = end - start
print("Time take : %f seconds" % elapsed)
print("Best parameters set found on development set:")
print()
print(clf.best_estimator_)
print()
print("Grid scores on development set:")
print()
for params, mean_score, scores in clf.grid_scores_:
print("%0.3f (+/-%0.03f) for %r" %
(mean_score, scores.std() / 2, params))
print()
print("Detailed classification report:")
print()
print("The model is trained on the full development set.")
print("The scores are computed on the full evaluation set.")
print()
y_true, y_pred = labelsTest, clf.predict(dataTest)
print(classification_report(y_true, y_pred))
print()
with open(os.path.join(modelDir, model + '_svm.pkl'), 'wb') as fid:
cPickle.dump(clf, fid)
def classify_rfc(feature,labels,model='left'):
print "---------Random Forest Classifier-------------"
modelDir = os.path.join(projectPath,model)
utility.checkDirectory(modelDir)
dataTrain,dataTest,labelsTrain,labelsTest = train_test_split(feature,
labels, test_size=0.20,
random_state=42)
param_grid = [
{'n_estimators': [1, 10, 100, 1000], 'max_features': [10,50,100, 400]},
]
rfc = RandomForestClassifier(n_estimators=10)
clf = grid_search.GridSearchCV(estimator=rfc, param_grid=param_grid,cv=5,n_jobs=-2)
print "Classification for "+model+" eye\n"
print "Training RFC classifier for grid of C and gamma values to select best parameter\n"
start = time.time()
clf.fit(dataTrain,labelsTrain)
end = time.time()
elapsed = end - start
print("Time take : %f seconds"%elapsed)
print("Best parameters set found on development set:")
print()
print(clf.best_estimator_)
print()
print("Grid scores on development set:")
print()
for params, mean_score, scores in clf.grid_scores_:
print("%0.3f (+/-%0.03f) for %r"
% (mean_score, scores.std() / 2, params))
print()
print("Detailed classification report:")
print()
print("The model is trained on the full development set.")
print("The scores are computed on the full evaluation set.")
print()
y_true, y_pred = labelsTest, clf.predict(dataTest)
print(classification_report(y_true, y_pred))
print()
with open(os.path.join(modelDir,model+'_5_rfc.pkl'), 'wb') as fid:
cPickle.dump(clf, fid)