关于dive_into_keras_vgg16的预测脚本问题

[littlecangbaby]

你好，目录下的ex.py文件可以跑过，但是后面预测的那部分源码在哪里，感觉看的不是很懂。能否放上完整的脚本供参考，非常感谢。

现在我们开始来预测了

首先写一个方法来加载并处理图片

def load_image(imageurl):
    im = cv2.resize(cv2.imread(imageurl),(224,224)).astype(np.float32)
    im[:,:,0] -= 103.939
    im[:,:,1] -= 116.779
    im[:,:,2] -= 123.68
    im = im.transpose((2,0,1))
    im = np.expand_dims(im,axis=0)
    return im

读取vgg16的类别文件

f = open('synset_words.txt','r')
lines = f.readlines()
f.close()

def predict(url):
    im = load_image(url)
    pre = np.argmax(model.predict(im))
    print lines[pre]

%pylab inline

Populating the interactive namespace from numpy and matplotlib

from IPython.display import Image

Image('cat1.jpg')

[Ontheroad123]

def load_image(url):
im = cv2.resize(cv2.imread(url), (224, 224)).astype(np.float32)
im[:,:,0] -= 103.939
im[:,:,1] -= 116.779
im[:,:,2] -= 123.68
im = im.transpose((2,0,1))
im = np.expand_dims(im, axis=0)
return im

def predict(url):
# 预测
f = open('synset_words.txt', 'r')
lines = f.readlines()
im = load_image(url)
pre = np.argmax(model.predict(im))
print(lines[pre])
f.close()

if name == "main":

# Test pretrained model
model = VGG_16('vgg16_weights.h5')
sgd = SGD(lr=1e-6, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(optimizer=sgd, loss='categorical_crossentropy')
predict('cat.1.jpg')