我目前正在处理Smiles数据集，并尝试使用深度学习来检测笑容是正面还是负面。我使用的机器是Raspberry Pi 3，运行的Python版本是3.7（不是2.7）。
我的训练集中总共有13165张图片。我希望将这些图片存储在一个数组中。然而，我遇到了一个问题，就是无法为形状为(13165, 32, 32, 3)的数组分配内存。
以下是源代码（shallownet_smile.py）：
from sklearn.preprocessing import LabelBinarizerfrom sklearn.model_selection import train_test_splitfrom sklearn.metrics import classification_reportfrom pyimagesearch.preprocessing import ImageToArrayPreprocessorfrom pyimagesearch.preprocessing import SimplePreprocessorfrom pyimagesearch.datasets import SimpleDatasetLoaderfrom pyimagesearch.nn.conv.shallownet import ShallowNetfrom keras.optimizers import SGDfrom imutils import pathsimport matplotlib.pyplot as pltimport numpy as npimport argparseap = argparse.ArgumentParser()ap.add_argument("-d", "--dataset", required=True, help="path to input dataset")args = vars(ap.parse_args())# grab the list of images we'll be describingprint("[INFO] loading images...")imagePaths = list(paths.list_images(args["dataset"]))sp = SimplePreprocessor(32, 32)iap = ImageToArrayPreprocessor()sdl = SimpleDatasetLoader(preprocessors=[sp, iap])(data, labels) = sdl.load(imagePaths, verbose=1)# convert values to between 0-1data = data.astype("float") / 255.0# partition our data into training and test sets(trainX, testX, trainY, testY) = train_test_split(data, labels, test_size=0.25, random_state=42)# convert the labels from integers to vectorstrainY = LabelBinarizer().fit_transform(trainY)testY = LabelBinarizer().fit_transform(testY)# initialize the optimizer and modelprint("[INFO] compiling model...")# initialize stochastic gradient descent with learning rate of 0.005opt = SGD(lr=0.005)model = ShallowNet.build(width=32, height=32, depth=3, classes=2)model.compile(loss="categorical_crossentropy", optimizer=opt, metrics=["accuracy"])# train the networkprint("[INFO] training network...")H = model.fit(trainX, trainY, validation_data=(testX, testY), batch_size=32, epochs=100, verbose=1)print("[INFO] evaluating network...")predictions = model.predict(testX, batch_size=32)print(classification_report( testY.argmax(axis=1), predictions.argmax(axis=1), target_names=["positive", "negative"]))plt.style.use("ggplot")plt.figure()plt.plot(np.arange(0, 100), H.history["loss"], label="train_loss")plt.plot(np.arange(0, 100), H.history["val_loss"], label="val_loss")plt.plot(np.arange(0, 100), H.history["acc"], label="train_acc")plt.plot(np.arange(0, 100), H.history["val_acc"], label="val_acc")plt.title("Training Loss and Accuracy")plt.xlabel("Epoch #")plt.ylabel("Loss/Accuracy")plt.legend()plt.show()假设数据集在我当前目录下。以下是我得到的错误信息：
python3 shallownet_smile.py -d=datasets/Smiles
我仍然感到困惑，不知道哪里出了问题。我非常希望有经验的深度学习/机器学习专家能够解释并澄清我做错了什么。
感谢您的帮助和关注。
回答：
首先，您的系统内存非常低，因此请尝试使用较小的图片。
错误主要源自这一行代码 data = data.astype("float") / 255.0
原因是数据已经是一个uint8类型的numpy数组，而您又在创建一个float32类型的numpy数组，这会占用额外的内存。
我将修改一些部分的simpledataloader，以便您可以进行训练。
转到 from pyimagesearch.datasets import SimpleDatasetLoader。它应该在文件夹pyimagesearch/datasets/simpledatasetloader.py中（示例代码：https://github.com/whydna/Deep-Learning-For-Computer-Vision/blob/master/pyimagesearch/datasets/simpledatasetloader.py）
用我的代码更改此.py文件，并更改max_image的值（除非您能处理现有的内存，否则请减少它），还要删除这一行 data = data.astype("float") / 255.0，因为我从函数中发送了预处理后的数组。
# import the necessary packagesimport numpy as npimport cv2import osmax_image = 1000class SimpleDatasetLoader: def __init__(self, preprocessors=None): # store the image preprocessor self.preprocessors = preprocessors # if the preprocessors are None, initialize them as an # empty list if self.preprocessors is None: self.preprocessors = [] def load(self, imagePaths, verbose=-1): # initialize the list of features and labels data = [] labels = [] cnt = 0 # loop over the input images for (i, imagePath) in enumerate(imagePaths): if cnt >= max_image: break # load the image and extract the class label assuming # that our path has the following format: # /path/to/dataset/{class}/{image}.jpg image = cv2.imread(imagePath) label = imagePath.split(os.path.sep)[-2] # check to see if our preprocessors are not None if self.preprocessors is not None: # loop over the preprocessors and apply each to # the image for p in self.preprocessors: image = p.preprocess(image) # treat our processed image as a "feature vector" # by updating the data list followed by the labels data.append(image) labels.append(label) # show an update every `verbose` images cnt += 1 if verbose > 0 and i > 0 and (i + 1) % verbose == 0: print("[INFO] processed {}/{}".format(i + 1, len(imagePaths))) # return a tuple of the data and labels return (np.array(data, dtype='float32')/255., np.array(labels))如果您仍然遇到内存问题，请在此处减少batch_size
H = model.fit(trainX, trainY, validation_data=(testX, testY), batch_size=4, epochs=100, verbose=1)
