Question

我正在尝试对图像识别进行编码，以针对不同的动物图像训练系统，这就是代码。我正在使用anaconda作为解释器，为环境使用pycharm。

import tensorflow as tf
import os, sys
from PIL import Image

image_path = 'test_images/leopard2.jpg'

size = (299, 299)

infile = image_path
outfile = os.path.splitext(infile)[0] + '_resized.jpg'
try:
  im = Image.open(infile)
  im.thumbnail(size, Image.ANTIALIAS)
  old_im_size = im.size

## By default, black colour would be used as the background for padding!
new_im = Image.new("RGB", size)

new_im.paste(im,(int(size[0]-old_im_size[0])/2,int(size[1]- 
old_im_size[1])/2))

new_im.save(outfile, "JPEG")
except IOError:
print("Cannot resize '%s'") %infile



# Read in the image_data
image_data = tf.gfile.FastGFile(outfile, 'rb').read()

# Loads label file, strips off carriage return
label_lines = [line.rstrip() for line 
                in tf.gfile.GFile("output_labels.txt")]

# Unpersists graph from file
with tf.gfile.FastGFile("output_graph.pb", 'rb') as f:
   graph_def = tf.GraphDef()
   graph_def.ParseFromString(f.read())
  _  = tf.import_graph_def(graph_def, name='')

init_ops = tf.global_variables_initializer()
with tf.Session() as sess:
sess.run(init_ops)
# Feed the image_data as input to the graph and get first prediction
softmax_tensor = sess.graph.get_tensor_by_name('final_result:0')

predictions = sess.run(softmax_tensor, \
         {'DecodeJpeg/contents:0': image_data})

# Sort to show labels of first prediction in order of confidence
top_k = predictions[0].argsort()[-len(predictions[0]):][::-1]

 for node_id in top_k:
    human_string = label_lines[node_id]
    score = predictions[0][node_id]
    print('%s (score = %.5f)' % (human_string, score))
 os.remove(outfile)

出现的错误是

C:\Users\snklp\Anaconda3\envs\untitled\python.exe 
C:/Users/snklp/Downloads/Transfer-Learning-for-Animal-Classification-in- 
Tensorflow-master/Transfer-Learning-for-Animal-Classification-in-Tensorflow- 
master/test.py
Traceback (most recent call last):
File "C:/Users/snklp/Downloads/Transfer-Learning-for-Animal-Classification- 
in-Tensorflow-master/Transfer-Learning-for-Animal-Classification-in- 
Tensorflow-master/test.py", line 19, in <module>
new_im.paste(im,(int(size[0]-old_im_size[0])/2,int(size[1]- 
old_im_size[1])/2))
File "C:\Users\snklp\Anaconda3\envs\untitled\lib\site-packages\PIL\Image.py", 
line 1423, in paste
self.im.paste(im, box)
TypeError: integer argument expected, got float

Process finished with exit code 1

我正在使用的图像为jpeg格式，并且位于代码中定义的正确路径中。有人知道这是什么问题吗？

Answer 1

您的2除法不会被int（）全部包裹。如果您的(size[0]-old_im_size[0])或(size[1]- old_im_size[1])出现奇数，则代码将中断。

尝试一下：

new_im.paste(im,(int((size[0]-old_im_size[0])/2), int((size[1]- old_im_size[1])/2)))

Answer 2

您可以使用int（）环绕所获取的float对象，也可以使用//而不是/进行除法。看到这个：-

>>> t = 44/2
>>> t
22.0
>>> int(t)
22
>>> t = 44//2
>>> t
22

预期为整数，浮点数

2 个答案: