从 CNN（Xception 模型）中提取特征，我得到 IsADirectoryError？

Question

从 xcption 模型中提取特征 我正在使用 cnn 和 lstm 开发图像标题生成器，并且我分享了从网站获得的代码。我试图解决它，但是当我对代码进行更改时，我在其他部分出现错误，所以任何人都可以告诉我如何解决这个问题

def extract_features(directory):
        model = Xception( include_top=False, pooling='avg' )
        features = {}
        for img in tqdm(os.walk(directory)):
            filename = directory + "/" + img
            image = Image.open(filename)
            image = image.resize((299,299))
            image = np.expand_dims(image, axis=0)
            #image = preprocess_input(image)
            image = image/127.5
            image = image - 1.0

            feature = model.predict(image)
            features[img] = feature
        return features

#2048 feature vector
features = extract_features(dataset_images)
dump(features, open("features.p","wb"))

错误

TypeError                                 Traceback (most recent call last)
<ipython-input-8-5f1d4e4e3211> in <module>()
     16 
     17 #2048 feature vector
---> 18 features = extract_features(dataset_images)
     19 dump(features, open("features.p","wb"))

<ipython-input-8-5f1d4e4e3211> in extract_features(directory)
      3         features = {}
      4         for img in tqdm(os.walk(directory)):
----> 5             filename = directory + "/" + img
      6             image = Image.open(filename)
      7             image = image.resize((299,299))

TypeError: can only concatenate str (not "tuple") to str

Answer 1

• 在您的情况下，filename = directory + "/" + img 中的 img 是一个元组，directory 是一个字符串，因此它们不能被连接。

• 所以你已经循环通过img。在这种情况下，您可能会收到另一个错误，因为img 包含根目录和子目录。

  def extract_features(directory):
              model = Xception( include_top=False, pooling='avg' )
              features = {}
              for (root,dirs,files) in tqdm(os.walk(directory)):
                  for img in files:
                      filename = directory + "/" + img
                      image = Image.open(filename)
                      image = image.resize((299,299))
                      image = np.expand_dims(image, axis=0)
                      #image = preprocess_input(image)
                      image = image/127.5
                      image = image - 1.0
                      feature = model.predict(image)
                      features[img] = feature
              return features