input_dataframe = pandas.read_csv(input_file)
value = np.array(input_dataframe['value'])
prediction = np.array(input_dataframe['prediction'])
prediction = np.array(input_dataframe['prediction'])
label = np.array(input_dataframe['label'])
prediction_training = np.array(input_dataframe['prediction_training'])
prediction_training_stops = 0
for i in range(0, len(prediction_training)):
if prediction_training[i] == 1:
prediction_training_stops += 1
testing_value = value[prediction_training_stops:]
testing_prediction = prediction[prediction_training_stops:]
detector_instance = detector.detector(values=testing_value,
predictions=testing_prediction)
warp_distance = detector_instance.calculate_distances(
comparision_window_size=dtw_window_size)
threshold = detector_instance.set_threshold(
training_ratio=training_ratio,
max_multipler=threshold_max_multipler)
positive_detection = detector_instance.get_anomalies()
threshold_training_starts = prediction_training_stops
threshold_training_size = int(len(testing_value) * training_ratio)
threshold_ignore = np.zeros(threshold_training_starts)
threshold_training = np.ones(threshold_training_size)
threshold_testing = np.zeros(
len(value) - threshold_training_starts - threshold_training_size)
threshold_training_colomn = np.append(threshold_ignore,
from utils import visualization_utils as vis_util
import numpy as np
import cv2
from detector import detector
from imutils import paths
import os
import time
det = detector.detector()
image_folder = 'data_voc'
#imagePaths = paths.list_images(image_folder)
imageNames = os.listdir(image_folder)
# loop over the image paths
for imageName in imageNames:
#print(imageName)
image = cv2.imread(image_folder + '/' + imageName)
print(image_folder + '/' + imageName)
h, w = image.shape[:2]
(boxes, scores, classes, num, category_index) = det.detect_plate(image)
# print(category_index)