Object-localization/test.py at master · tetsuyaAnkur/Object-localization · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
import model2
import preprocessQ2 as pp
import numpy as np
import os
import cv2
from keras.preprocessing.image import ImageDataGenerator, array_to_img, img_to_array, load_img

def training(a):
	x_train,y_train,x_test,y_test=pp.preprocess()
	x_train=x_train.astype(np.float32)
	x_test=x_test.astype(np.float32)
	y_train=y_train.astype(np.float32)
	y_test=y_test.astype(np.float32)
	x_train=x_train/255
	x_test=x_test/255
	print(x_train.shape,y_train.shape)
	a.load_weights('/home/arun/Desktop/AI_FINAL/VOCdevkit/voc2010/4c5dweights.h5')
	for x in x_test:
		l=[x]
		l=np.asarray(l)
		y=a.predict(l,batch_size=1)
		print(y)
		temp=x*255
		temp=temp.astype(np.uint8)
		cv2.rectangle(temp,(int(y[0][0]),int(y[0][2])),(int(y[0][1]),int(y[0][3])),(0,0,255),2)
		cv2.imshow('image',temp)
		cv2.waitKey(1000)
	print(a.evaluate(x_test,y_test,batch_size=32))
	return a

#model = resnet34.modeling(50,50,3)
'''
dir_path = '/home/ankur/VOCdevkit/VOC2010/Annotations'
dir_path2 = '/home/ankur/VOCdevkit/VOC2010/JPEGImages'
data = parsing(dir_path,dir_path2)
'''

model1=model2.modeling(500,500,3)
training(model1)