Question

我有以下Python脚本：

java version "9.0.1"
Java(TM) SE Runtime Environment (build 9.0.1+11)
Java HotSpot(TM) 64-Bit Server VM (build 9.0.1+11, mixed mode)

这是传递图像文件名及其标签的正确方法吗？如果import numpy as np from PIL import Image names = [] X = [] labels = [] with open('data.txt', 'r') as f: for line in f.readlines(): tokens = line.split(' ') names.append(tokens[0]) labels.append(int(tokens[1])) for img in range(len(names)): for name in names: img = np.array(Image.open(name)) X.append(img) X = np.array(X) X_to_text_file = np.savetxt('x.txt', X.reshape(np.shape(X)), fmt='%5f') X_data = import_data('x.txt')具有以下格式：

data.txt

我也在下面收到错误：img1.png 1 img2.png 1 img3.png 2 img4.png 1 img5.png 1

X_to_text_file = np.savetxt('x.txt', X.reshape(np.shape(X)), fmt='%5f')

有关于此的任何想法吗？

感谢。

Answer 1

演示：

from glob import glob
import cv2

In [284]: names = glob(r'D:\temp\photo\*.jpg')

In [285]: names
Out[285]:
['D:\\temp\\photo\\20081116-IMG_0900.jpg',
 'D:\\temp\\photo\\20081116-IMG_0902.jpg']

In [286]: X = np.array([cv2.imread(f) for f in names])

In [287]: X.shape
Out[287]: (2, 2112, 2816, 3)

如何将训练数据传递给神经网络

1 个答案: