hand-writting num recognizer

/Test programme/: This programme is my first deepLearning programme.

This projekt is designed to recognize hand-writting number image.

深度学习:构建神经网络并识别手写数字

正确率:0.63764(好像有点低

正确率:0.91175(发现有个cell有点错误,修正后达到91%了,好耶!

疑难杂症

tenserflow把numpy卸载了…

没关系,会安装对应版本的numpy的

Python关于%matplotlib inline报错

这个是 jupyter notebook 或者 jupyter qtconsole 的专属代码,使用%matplotlib inline这段代码可以输出图像。但是vscode编辑器并没有这个功能。(但这并不妨碍vscode是宇宙第一编辑器的事实)

FileNotFoundError

PS F:\Programme_works\Python> python -u "f:\Programme_works\Python\DeepLearning\numRecog\neuralNetwork.py"
Traceback (most recent call last):
  File "f:\Programme_works\Python\DeepLearning\numRecog\neuralNetwork.py", line 106, in \<module\>
    with open('train.csv', 'r') as data_file:
FileNotFoundError: [Errno 2] No such file or directory: 'train.csv'

未找到文件,检查一下文件是否在命令执行时的目录(一般都是vscode打开的文件夹,或者jupyter notebook运行的文件所在的文件夹)

非jupyter编辑器输出csv图像

matplotlib.pyplot.imshow(image_array, cmap='Greys', interpolation='None')

改为

matplotlib.pyplot.imsave('temp.png', image_array, cmap='Greys')

便会保存在目录下