系统学习：使用OpenCV和Python实现的机器学习

欢迎加入全球最大AI社群>>

前言

https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/00.00-Preface.ipynb

https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/00.01-Foreword-by-Ariel-Rokem.ipynb

1. 机器学习的味道

   https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/01.00-A-Taste-of-Machine-Learning.ipynb
   

2. 在OpenCV中使用数据

   https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/02.00-Working-with-Data-in-OpenCV.ipynb
   

• 使用Python的NumPy软件包处理数据

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/02.01-Dealing-with-Data-Using-Python-NumPy.ipynb
  

• 在Python中加载外部数据集

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/02.02-Loading-External-Datasets-in-Python.ipynb
  

• 使用Matplotlib可视化数据

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/02.03-Visualizing-Data-Using-Matplotlib.ipynb
  

• 使用OpenCV的TrainData容器处理数据

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/02.05-Dealing-with-Data-Using-the-OpenCV-TrainData-Container-in-C%2B%2B.ipynb
  

3. 监督学习的第一步

   https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/03.00-First-Steps-in-Supervised-Learning.ipynb
   

• 用评分功能测量模型性能

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/03.01-Measuring-Model-Performance-with-Scoring-Functions.ipynb
  

• 了解k-NN算法

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/03.02-Understanding-the-k-NN-Algorithm.ipynb
  

• 使用回归模型预测持续成果

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/03.03-Using-Regression-Models-to-Predict-Continuous-Outcomes.ipynb
  

• 应用拉索和岭回归

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/03.04-Applying-Lasso-and-Ridge-Regression.ipynb
  

• 使用Logistic回归分类虹膜物种

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/03.05-Classifying-Iris-Species-Using-Logistic-Regression.ipynb
  

4. 代表数据和工程特性

   https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/04.00-Representing-Data-and-Engineering-Features.ipynb
   

• 预处理数据

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/04.01-Preprocessing-Data.ipynb
  

• 减少数据的维度

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/04.02-Reducing-the-Dimensionality-of-the-Data.ipynb
  

• 代表分类变量

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/04.03-Representing-Categorical-Variables.ipynb
  

• 表示文本特征

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/04.04-Represening-Text-Features.ipynb
  

• 代表图像

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/04.05-Representing-Images.ipynb
  

5. 使用决策树进行医学诊断

   https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/05.00-Using-Decision-Trees-to-Make-a-Medical-Diagnosis.ipynb
   

• 建立你的第一决策树

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/05.01-Building-Your-First-Decision-Tree.ipynb
  

• 使用决策树诊断乳腺癌

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/05.02-Using-Decision-Trees-to-Diagnose-Breast-Cancer.ipynb
  

• 使用决策树回归

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/05.03-Using-Decision-Trees-for-Regression.ipynb
  

6. 用支持向量机检测行人

   https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/06.00-Detecting-Pedestrians-with-Support-Vector-Machines.ipynb
   

• 实施您的第一支持向量机

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/06.01-Implementing-Your-First-Support-Vector-Machine.ipynb
  

• 检测野外行人

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/06.02-Detecting-Pedestrians-in-the-Wild.ipynb
  

• 附加SVM练习

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/06.03-Additional-SVM-Exercises.ipynb
  

7. 用贝叶斯学习实现垃圾邮件过滤器

   https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/07.00-Implementing-a-Spam-Filter-with-Bayesian-Learning.ipynb
   

• 实现我们的第一个贝叶斯分类器

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/07.01-Implementing-Our-First-Bayesian-Classifier.ipynb
  

• 分类电子邮件使用朴素贝叶斯

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/07.02-Classifying-Emails-Using-Naive-Bayes.ipynb
  

8. 用无监督学习发现隐藏的结构

   https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/08.00-Discovering-Hidden-Structures-with-Unsupervised-Learning.ipynb
   

• 了解k均值聚类

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/08.01-Understanding-k-Means-Clustering.ipynb
  

• 使用k-Means压缩彩色图像

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/08.02-Compressing-Color-Images-Using-k-Means.ipynb
  

• 使用k-Means分类手写数字

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/08.03-Classifying-Handwritten-Digits-Using-k-Means.ipynb
  

• 实施聚集层次聚类

• https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/08.04-Implementing-Agglomerative-Hierarchical-Clustering.ipynb
  

9. 使用深度学习分类手写数字

   https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/09.00-Using-Deep-Learning-to-Classify-Handwritten-Digits.ipynb
   

• 了解感知器

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/09.01-Understanding-Perceptrons.ipynb
  

• 在OpenCV中实现多层感知器

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/09.02-Implementing-a-Multi-Layer-Perceptron-in-OpenCV.ipynb
  

• 认识深度学习

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/09.03-Getting-Acquainted-with-Deep-Learning.ipynb
  

• 在OpenCV中培训MLP以分类手写数字

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/09.04-Training-an-MLP-in-OpenCV-to-Classify-Handwritten-Digits.ipynb
  

• 训练深层神经网络使用Keras分类手写数字

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/09.05-Training-a-Deep-Neural-Net-to-Classify-Handwritten-Digits-Using-Keras.ipynb
  

10. 将不同的算法合并成一个合奏

    https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/10.00-Combining-Different-Algorithms-Into-an-Ensemble.ipynb
    

• 了解组合方法

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/10.01-Understanding-Ensemble-Methods.ipynb
  

• 将决策树组合成随机森林

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/10.02-Combining-Decision-Trees-Into-a-Random-Forest.ipynb
  

• 使用随机森林进行人脸识别

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/10.03-Using-Random-Forests-for-Face-Recognition.ipynb
  

• 实施AdaBoost

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/10.04-Implementing-AdaBoost.ipynb
  

• 将不同的模型组合成投票分类器

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/10.05-Combining-Different-Models-Into-a-Voting-Classifier.ipynb
  

11. 使用超参数调整选择正确的模型

    https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/11.00-Selecting-the-Right-Model-with-Hyper-Parameter-Tuning.ipynb
    

• 评估模型

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/11.01-Evaluating-a-Model.ipynb
  

• 了解交叉验证，Bootstrapping和McNemar的测试

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/11.02-Understanding-Cross-Validation-Bootstrapping-and-McNemar's-Test.ipynb
  

• 使用网格搜索调整超参数

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/11.03-Tuning-Hyperparameters-with-Grid-Search.ipynb
  

• 链接算法一起形成管道

  https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/11.04-Chaining-Algorithms-Together-to-Form-a-Pipeline.ipynb
  

12. 结束语

    https://github.com/mbeyeler/opencv-machine-learning/blob/master/notebooks/12.00-Wrapping-Up.ipynb

热门文章推荐

重磅｜腾讯公布首批25个ai加速器项目名单！

国家战略：国务院发布《新一代人工智能发展规划》

京东发布登月机器学习平台：为第四次零售革命输出AI能力

重磅｜中国870家ai公司融资905亿人民币！

浙大女科学家解密：从最后一面逆袭第一名！

资料｜麻省理工课程：深度学习数据基础（ppt)

推荐｜40张动态图详解全部传感器关注原理！

警惕!中国人工智能有一只推荐算法叫:莆田系算法！

百度Apollo：无人驾驶技术发展成熟仅需3年左右！

阿里出了个Take Go无人便利店，比亚马逊还厉害！