def main(testfile=None): if testfile: a=rd.read(testfile) k=3 predictions=[] for x in range(len(a)): neighbors = knn.getNeighbors(train, a[x], k) result = knn.getResponse(neighbors) predictions.append(result) return(predictions) else: knn.main(train,tests)
def main(i=0, j=62, testfile=None):
tr, ts = rt.read2(i, j)
k = 3
if testfile:
tst = rd.read2(testfile)
predictions = []
for x in range(len(tst)):
neighbors = knn.getNeighbors(tr, tst[x], k)
result = knn.getResponse(neighbors)
predictions.append(result)
return (predictions)
else:
knn.main(tr, ts)
def getClassificationPrecision(trainingSet,testSet,k=3):
predictions=[]
preds=[[0,0,0,0,0],[0,0,0,0,0],[0,0,0,0,0],[0,0,0,0,0],[0,0,0,0,0]]
dictClasses={'carettacaretta': 0, 'cheloniamydas': 1,'dermochelyscoriacea':2,'eretmochelysimbricata':3,'lepidochelysolivacea':4}
for x in range(len(testSet)):
neighbors = knn.getNeighbors(trainingSet, testSet[x], k)
result = knn.getResponse(neighbors)
#print result
predictions.append(result)
indiceClassePred=dictClasses[result]
indiceClasseActual=dictClasses[testSet[x][-1]]
preds[indiceClasseActual][indiceClassePred]=preds[indiceClasseActual][indiceClassePred]+1
#print('> predicted=' + repr(result) + ', actual=' + repr(testSet[x][-1]))
#print predictions
precisao = knn.getPrecision(testSet, predictions)
#print('Accuracy: ' + repr(accuracy) + '%')
#quit()
return precisao
split = 0.67
print 'Train set: ' + repr(len(trainingSet))
print 'Test set: ' + repr(len(testSet))
# generate predictions
predictions=[]
k = 3 # valor de k igual a 3
preds=[[0,0,0,0,0],[0,0,0,0,0],[0,0,0,0,0],[0,0,0,0,0],[0,0,0,0,0]]
dictClasses={'carettacaretta': 0, 'cheloniamydas': 1,'dermochelyscoriacea':2,'eretmochelysimbricata':3,'lepidochelysolivacea':4}
numeroImagensPorClasse=30.00;
for x in range(len(testSet)):
neighbors = knn.getNeighbors(trainingSet, testSet[x], k)
result = knn.getResponse(neighbors)
predictions.append(result)
indiceClassePred=dictClasses[result]
indiceClasseActual=dictClasses[testSet[x][-1]]
preds[indiceClasseActual][indiceClassePred]=preds[indiceClasseActual][indiceClassePred]+1
print('> predicted=' + repr(result) + ', actual=' + repr(testSet[x][-1]))
accuracy = knn.getAccuracy(testSet, predictions)
print('Accuracy: ' + repr(accuracy) + '%')
im1, tests_contours, tests_hierarchy = cv2.findContours(
closed_tests, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
tests_feature = createFeature(tests_contours, "")
cv2.drawContours(tests_draw, tests_contours, -1, (255, 0, 0), 3)
cv2.imshow("tests - contours", tests_draw)
cv2.waitKey(0)
cv2.destroyAllWindows()
k = 5
all_feature = nuts_feature + bolts_feature
#for test tests_contours[0]
testInstance = [tests_feature[0][0], tests_feature[0][1]]
neighbors = knn.getNeighbors(all_feature, testInstance, k)
height = 500
width = 500
feature_space = np.zeros((height, width, 3), np.uint8)
for i in range(len(all_feature)):
if all_feature[i][2] == 'nut':
cv2.circle(
feature_space,
(int(all_feature[i][0] * width), int(all_feature[i][1] * height)),
2, (0, 0, 255), -1)
else:
cv2.circle(
feature_space,
(int(all_feature[i][0] * width), int(all_feature[i][1] * height)),
accuracy_lvq3_k3 = []
accuracy_knn = []
for n_prototypes in number_prototypes:
prototypes_lvq1 = train_prototypes_lvq1(lvq_training_set, n_prototypes,
lrate, epochs)
prototypes_lvq2 = train_prototypes_lvq2(prototypes_lvq1,
lvq_training_set, lrate,
epochs)
prototypes_lvq3 = train_prototypes_lvq3(prototypes_lvq2,
lvq_training_set, lrate,
epochs)
for k in kn:
predictions = []
for row in range(len(knn_test_set)):
neighbors = getNeighbors(prototypes_lvq1, knn_test_set[row], k)
results = getResponse(neighbors)
predictions.append(results)
accuracy = getAccuracy(knn_test_set, predictions)
if k == 1:
accuracy_lvq1_k1.append(accuracy)
else:
accuracy_lvq1_k3.append(accuracy)
print('With LVQ1 for ' + str(n_prototypes) +
' and for dataset -> ' + str(filename) +
' accuracy for k= ' + str(k) + ': ' + repr(accuracy) + '%')
for k in kn:
predictions = []
for row in range(len(knn_test_set)):
neighbors = getNeighbors(prototypes_lvq2, knn_test_set[row], k)
