资源列表
SVM做图片处理
- 使用SVM算法对CIFAR-10图片数据集进行分类，包括模型的训练，测试和参数的调优(Using SVM algorithm to classify CIFAR-10 image data sets, including model training, testing and parameter tuning)
DNN
- 利用python3完整实现DNN，包括前向传播和反向传播。实现一个2次函数的拟合。(Complete implementation of DNN using python3, including forward propagation and reverse propagation. Implement a quadratic function fitting.)
模拟退火算法
- 使用模拟退火算法对Bp神经网络进行改进，效果比较明显(Using simulated annealing algorithm to improve the Bp neural network, the effect is obvious.)
sa_bp
- 基于模拟退火算法对BP神经网络进行的改进，有效提高了精度和迭代速度。(The improvement of BP neural network based on simulated annealing algorithm effectively improves the accuracy and speed of iteration.)
yolo_tensorflow
- 基于YOLO的目标检测, 对十种物体进行识别和定位。(object detection based on YOLO, perform recognition and location to ten types of objects.)
TransferEntropy1
- 传递熵算法代码，用于分析不同波动之间的关系，包括时延、权重等仅供参考(Transfer Entropy Algorithm Code, used to analyze the relationship between different fluctuations, including delay, weight, etc. for reference only)
Ch04
- 贝叶斯算法使用python ,内含测试数据文件，数据为多个txt格式的邮件文件。(The bayesian algorithm USES python, which contains test data files that are multiple mail files in TXT format.)
Chapter04
- 基于tensorflow 的神经网络的损失函数，学习率，正则化，滑动平均等方法(Method of loss function, learning rate, regularization and sliding average of neural network based on tensorflow)
基于SOM的数据分类
- SOM神经网络也属于自组织型学习网络，只不过更特殊一点它属于自组织特征的映射网络。该网络是由一个全连接的神经元阵列组成的无教师，自组织，自学习的网络。(SOM neural network also belongs to self-organizing learning network, but more specifically, it belongs to self-organizing feature mapping network. The network is a non-teache
7-9
- 分别介绍了概率与数理统计概述、统计估计、假设检验、方差分析、回归分析、正交试验分析、聚类分析、判别分析和多元数据相关分析等内容，理论与实践相结合，向读者演示了matlab在数理统计中的应用。(It respectively introduces the outline of probability and mathematical statistics, statistical estimation, hypothesis test, variance analysis, regression
dskpla
- MATLAB的函数、算法基本库和实例。 里面包含大量的运算函数，基本可以直接调用使用，提高效率(MATLAB function, algorithm base library and examples. It contains a lot of operation functions, which can be called directly to improve efficiency)
bnn-master
- 一个高度优化的轻量深度学习前向框架，使用C/C++语言开发，跨平台，支持读取Caffe模型文件，主要处理卷积神经网络。与市面上大多数移动端解决方案不同，我们的量化压缩技术不仅针对模型的权重，还涉及到输入的特征向量压缩。针对这一特性我们在模型文件和内存大小得到裁剪的同时还对框架的性能做了大量优化。(A highly optimized lightweight deep learning forward framework, developed using C/C++ language, cross