搜索资源列表
Extended Yale B Database
- 这是MNIST数据库(一个手写数字的数据库,它提供了六万的训练集和一万的测试集,它的图片是被规范处理过的,28*28的灰度图) 总共4个文件: train-labels-idx1-ubyte: training set labels t10k-images-idx3-ubyte:? test set images t10k-labels-idx1-ubyte:? test set labels train-images-idx3-ubyte: training set images(T
纯C-CNN
- 纯C深度学习库,里面包含MNIST手写数字识别数据集,编译就能训练和预测(Pure C depth learning library, which contains MNIST handwritten digital recognition data sets, compiling can be trained and predicted.)
lib.tar
- 将训练之后的caffe model生成.so库文件,方便在工程中直接调用,不需要写caffe test分代码(he Caffe model after the training is generated for the.so library file, which is convenient to call in the project and does not need to write Caffe test code)
chapter19
- 在MATLAB平台上的基于svm的手写数字体识别(Handwritten numeral recognition based on svm)
python-dbn-master
- 运用python语言,基于dbn的手写数字体识别(Handwritten numeral recognition based on dbn using python language)
Run_MNIST
- 下载MNIST数据集(手写体数字0-9)后,搭建卷积神经网络,将输入的数据集经过一层一层的卷积,到最后计算交叉熵,用梯度下降算法去优化它,使它变得最小,这就训练出了权重和偏置量,识别的准确率为91%(Download the MNIST data set (handwritten number 0-9), build a convolutional neural network, the input data set by convolutional layers, finally calcul
my_cnn.tar
- 用卷积神经网络实现手写数字识别,数据集为mnist数据集(Convolution neural network is used to realize handwritten numeral recognition. Data set is MNIST data set.)
MNLIST and CNN
- 实现了在Mnist上的分类,使用了卷积神经网络(use convoluntional neural network to implement classificaiton on Minist.)
mnist
- 利用keras实现手写数字识别,使用CNN模型 全连接层+两个卷积层,最后Softmax分类器,识别率超过96%(Using keras to realize handwritten numeral recognition baesd on CNN model. One whole connection layer + two convolution layers, and a Softmax classifier. The recognition accuracy is over 96%
rasmusbergpalm-DeepLearnToolbox-5df2801
- mnist数据库,可用matlab运行,学习神经网络(MNIST database can be run by MATLAB, learning neural network.)
dropout_and_minibatch
- 基于两层BP神经网络,加入dropout和softmax,输出层使用softmax,实现对手写字符库MNIST的识别,正确率达90%。(Based on the two level BP neural network, adding dropout and softmax, the output layer uses softmax to realize the recognition of handwritten character library MNIST, the accuracy ra
mnist1
- 改进的dnn,准确率达到了百分之95,有注释(Improved DNN, the accuracy rate reached ninety-five percent)
mnist98
- 改进的dnn,准确率达到了百分之98,有注释(The accuracy of the improved DNN is ninety-eight percent, with annotations)
mnistcnn
- 改进的cnn,解决了有的人跑起来显卡内存不足的问题(The improved CNN solves the problem of some people running up and displaying the card's memory.)
CNN
- 手写体识别的训练,采用卷积神经网络,附带数据集下载代码(The training of handwritten recognition is based on convolution neural network, and the download from the dataset.)
pytorch-vae-master
- 变分子编码 重构图像 Mnist 特征提取(vae reconstruction Mnist feature extracting)
tensorflow-mnist-predict-master
- 这个项目由四个脚本组成: 1. _create_model_1.py_ - 基于初学者教程创建一个模型model.ckpt文件。 2. * create_model_2.py * - 基于专家教程创建模型model2.ckpt文件。 3. * predict_1.py * - 使用model.ckpt(初学者教程)文件来预测.png文件中手写数字的正确整数。 4. * predict_2.py * - 使用model2.ckpt(专家教程)文件来预测.png文件中手写数字的正确整数。
CNN_SAR
- MNIST手写体识别,编程语言为matlab(MNIST handwriting recognition)
dbn_tf-master
- 利用深度置信网络实现对mnist数据集的分类(sort out the set of mnist with DBN)
mnist
- 使用了全连接网络,卷积神经网络,循环神经网络分别构建不同的分类器,如何通过模型保存原理进行保存。(Using the fully connected network and convolution neural network, recurrent neural network builds different classifiers respectively, and how to save them through the preservation principle of the mode