问题描述
目前我正在学习神经网络，我正在尝试创建一个可以训练识别手写字符的应用程序.对于这个问题，我使用了一个前馈神经网络，当我训练它识别 1、2 或 3 个不同的字符时，它似乎可以工作.但是当我尝试让网络学习超过 3 个字符时，它会停滞在 40 - 60% 左右的错误百分比.
Currently I'm learning about neural networks and I'm trying to create an application that can be trained to recognize handwritten characters.For this problem I use a feed-forward neural network and it seems to work when I train it to recognize 1, 2 or 3 different characters. But when I try to make the network learn more than 3 characters it will stagnate at a error percentage around the 40 - 60%.
我尝试了多层和更少/更多的神经元，但我似乎无法做到正确，现在我想知道前馈神经网络是否能够识别这么多信息.
I tried with multiple layers and less/more neurons but I can't seem to get it right, now I'm wondering if a feedforward neural network is capable of recognizing that much information.
一些统计数据:
网络类型:前馈神经网络
输入神经元: 100 (a 10 * 10) 网格用于绘制字符
Input neurons: 100 (a 10 * 10) grid is used to draw the characters
输出神经元:要重新识别的字符数量
有谁知道我的架构中可能存在的缺陷是什么?输入神经元是否过多?前馈神经网络是不是不能进行字符识别?
推荐答案
对于手写字符识别你需要
For handwritten character recognition you need
- 许多训练示例(也许您应该扭曲训练集)
- 输出层中的softmax激活函数
- 交叉熵误差函数
- 使用随机梯度下降训练
- 每一层的偏差
一个很好的测试问题是手写数字数据集MNIST.以下是在该数据集上成功应用神经网络的论文:
A good test problem is the handwritten digit data set MNIST. Here are papers that successfully applied neural networks on this data set:
是的.LeCun、L. Bottou、Y. Bengio 和 P. Haffner:基于梯度的学习应用于文档识别，http://yann.lecun.com/exdb/publis/pdf/lecun-98.pdf
Y. LeCun, L. Bottou, Y. Bengio and P. Haffner: Gradient-Based Learning Applied to Document Recognition, http://yann.lecun.com/exdb/publis/pdf/lecun-98.pdf
Dan Claudiu Ciresan、Ueli Meier、Luca Maria Gambardella、Juergen Schmidhuber:Deep Big Simple Neural Nets Excel on Handwritten Digit Recognition, http://arxiv.org/abs/1003.0358
Dan Claudiu Ciresan, Ueli Meier, Luca Maria Gambardella, Juergen Schmidhuber: Deep Big Simple Neural Nets Excel on Handwritten Digit Recognition, http://arxiv.org/abs/1003.0358
我训练了一个具有 784-200-50-10 架构的 MLP，在测试集上的准确率超过了 96%.
I trained an MLP with 784-200-50-10 architecture and got >96% accuracy on the test set.
这篇关于为 OCR 训练前馈神经网络的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！