def convert_file():
# for i in range(len(files)):
for i in range(1):
file_x = open((FOLDER + files[i]), 'r')
if not exists(SAVE_FOLDER):
mkdir(SAVE_FOLDER)
training_file = open((SAVE_FOLDER + "TR - " + files[i]), 'w')
test_file = open((SAVE_FOLDER + "TE - " + files[i]), 'w')
count_file = open((SAVE_FOLDER + "N - " + files[i]), 'w')
lines_x = file_x.readlines()
d, p, s, h, r = 0, 0, 0, 0, 0
test = lines_x[0].split(",")
w_line = []
w_line.append(str(range(0, len(test) - 1)).replace("[", "").replace("]", "") + ", Phase\n")
for index in range(len(lines_x)):
# print(lines_x[index], " " + str(len(lines_x[index])), " ", lines_x[index][len(lines_x[index]) - 2:len(lines_x[index]) - 1])
char = lines_x[index][len(lines_x[index]) - 2:len(lines_x[index]) - 1]
if char == "D":
d += 1
if char == "P":
p += 1
if char == "S":
s += 1
if char == "H":
h += 1
if char == "R":
r += 1
w_line.append(lines_x[index].replace(char, Pds.convert_target_name(char)))
count_line = ("D = " + str(d) + "\n" + "P = " + str(p) + "\n" + "S = " + str(s) + "\n" + "H = " + str(h) + "\n" + "R = " + str(r))
count_file.write(count_line)
count_file.close()
for index in range(0, (int(len(w_line) * 0.7))):
training_file.write(w_line[index])
training_file.close()
for index in range(int(len(w_line) * 0.7), len(w_line)):
test_file.write(w_line[index])
test_file.close()
hoopPos = np.zeros((2, 2), np.int)
# hoopPos = LabelFunc.getHoopPosition(label_vFn)
# hoopPos = np.array([[865, 79], [965, 179]]) # ball1
# hoopPos = np.array([[505, 272], [555, 322]]) # ball3
if task == 'label':
LabelFunc.labelGoalFrames(label_vFn, hoopPos, label_vAnnFile)
elif task == 'crop':
LabelFunc.cropHoop(crop_vFn, hoopPos, crop_vAnnFile, outDIRPos,
outDIRNeg)
elif task == 'compose':
PictureCombine.image_compose(outDIRPos, int(crop_size), int(crop_row),
int(crop_column), crop_pos_save)
PictureCombine.image_compose(outDIRNeg, int(crop_size), int(crop_row),
int(crop_column), crop_neg_save)
PrepareDataSet.select_test_img(crop_pos_save, crop_neg_save, test_path,
train_pkl, test_pkl)
elif task == 'training':
hog_center = np.mean(HogFeature.compute_hog(train_path), 0)
HogFeature.test_distance(hog_center, center_pos_path,
pos_distance_path)
HogFeature.test_distance(hog_center, center_neg_path,
neg_distance_path)
HogFeature.test_distance(hog_center, test_path, test_distance_path)
elif task == 'ROC':
pos_distance = ReadFile.load_distance(pos_distance_path)
neg_distance = ReadFile.load_distance(neg_distance_path)
threshold_values = ReadFile.get_threshold(pos_distance, neg_distance)
test_data = ReadFile.load_distance(test_distance_path)
test_img, test_label = ReadFile.load_pkl(test_pkl)
test_label = np.array(test_label, dtype=int)
ROC.plotROC(test_data, test_label, threshold_values)
y = var_exp
x = [i for i in range(len(eigen_values))]
soma = 0
for index in range(15):
soma += var_exp[index]
print soma
plt.plot(x, y, linestyle='--', marker='o', color='b')
plt.ylabel("Porcentagem de Representacao")
plt.xlabel("Indice dos Autovalores")
plt.show()
dataset = pds.get_dataset(pds.FILE)
reduced_matrix = execute(dataset)
print ("final", reduced_matrix)
# with open(pds.PATH+"/a1_va3_reducedR.csv", 'w') as csvw:
# csvw = csv.writer(csvw, delimiter=',')
# csvw.writerows(reduced_matrix)
np.savetxt(pds.FILE_REDUCED, reduced_matrix, delimiter=',', fmt='%.8f')
print('y', dataset['y'])
outf = open(pds.FILE_REDUCED_PRED, 'w')
for index in range(len(dataset['y'])):
