搜索资源列表
CNN
- 整理的有关卷积神经网络的介绍,算法原理,和一些关键代码(The introduction of convolutional neural network, algorithm principle, and some key code)
DeepLearning
- 包含神经网络、卷积神经网络、深度信念网络等深度学习程序,该程序可以用于语音识别、分类等处。(Deep learning programs including neural network, convolutional neural network and deep belief network can be used for speech recognition, classification and so on.)
cnn-mnist
- CNN-mnist自制算法,使用卷积神经网络进行计算,准确率99.2(CNN-mnist is a algorithm written by yourself.A convolution neural network is used for calculation, the accuracy rate is 99.2)
FaceRecognition_CNN(olivettifaces)
- 卷积神经网络面部识别利用python 实现(Python implementation of convolution neural network face recognition)
winter1516_assignment2
- 斯坦福大学CS231n《卷积神经网络》第二次作业代码,assignment2,(Standford CS231n second job code)
visual
- 实现了一个简单的卷积神经网络,并对卷积过程中的提取特征进行了可视化.(A simple convolution neural network is implemented, and the extraction features in the convolution process are visualized.)
nvdla-vmod
- 卷积神经网络实现的参考,借助于英伟达的开源硬件加速器。(Convolutional neural network reference, with open source hardware accelerator in nvidia.)
新建文件夹 (2)
- 本程序是用python写的一个深度学习程序,该程序实现了卷积神经网络(This procedure is written in python with a deep learning program, the program achieved a convolution neural network)
keras_mnist_test
- hello Word of keras ,第一个成功实现的卷积神经网络,下载了mnist数据集,并decode,,然后,为了加快速度,训练其中的一部分数据,并用predict测试,ok,2min内出结果.(网上其它程序试过,训练太久,一晚上都没训练出结果,于是自己动手设计了这个小程序) 环境:python3.6,keras2.1,PC i5(很破的电脑)(Hello Word of keras, the first successful convolution neural network,
CNN
- 卷积神经网络是近年发展起来,并引起广泛重视的一种高效识别方法。20世纪60年代,Hubel和Wiesel在研究猫脑皮层中用于局部敏感和方向选择的神经元时发现其独特的网络结构可以有效地降低反馈神经网络的复杂性,继而提出了卷积神经网络(Convolutional Neural Networks-简称CNN)。现在,CNN已经成为众多科学领域的研究热点之一,特别是在模式分类领域,由于该网络避免了对图像的复杂前期预处理,可以直接输入原始图像,因而得到了更为广泛的应用。(Convolution neura
CNN
- %得到卷积层的featuremap的size,卷积层fm的大小 为 上一层大小 - 卷积核大小 + 1(步长为1的情况) %得到卷积层的featuremap的size,卷积层fm的大小 为 上一层大小 - 卷积核大小 + 1(步长为1的情况)(Wang Yuting has a severe muscular dystrophy with only two arms to move about 5 centimeters. But she was determined to be a pain
LIFT-master
- 基于训练端对端的深度卷积神经网络学习局部不变性特征(Local invariant feature of learning images based on the training end to end convolution neural network)
gcForest-master
- 基于决策树构建深度森林模型实现较高特征表示能力相比深度卷积神经网络(Building deep forest model based on decision tree to achieve higher feature representation ability compared with deep convolution neural network)
faq1(1)
- cnn是卷积神经网络,代码利用cnn进行文本分类(CNN is a convolution neural network, and the code uses CNN for text classification)
3DCNN-master
- pointcloud 3D卷积神经网络(pointcloud 3DCNN)
MNIST_CNN
- 用于MNIST数据集,训练卷积神经网络,预测准确率大约为99.3%(Training Convolutional Neural Network on MNIST dataset)
卷积神经网络 - 副本
- 对LeNet-5的matlab实现,识别MINST手写数字集,压缩文件包含程序文件和训练、测试图片文件,程序可直接运行(版本matlab R2008a)(For the matlab implementation of LeNet-5, the MINST handwritten numeric set is identified. The uploaded file contains the program file and the training and test file. The pr
Lenet
- 经典五层卷积神经网络,应用在mnist库上(The classical five layer convolution neural network application in the MNIST database)
machine learning
- 反向传播算法与利用卷积神经网络识别手写体(Back propagation algorithm and recognition of handwriting by using convolution neural network)
NiftyNet-dev
- 医疗3d处理,即基于卷积神经网络的医疗影像分析平台(NiftyNet is a TensorFlow-based open-source convolutional neural networks (CNNs) platform for research in medical image analysis)