udacity machine learning

The process of building a machine learning algorithm: Quickly build a simple algorithm and test the performance of the algorithm with a cross-validation set. Draw the learning curve, check whether the algorithm has high variance or high deviation problem, so as to choose corresponding coping methods. Error analysis, to see the examples of errors

Here are just a few common machine learning methods. To understand the idea of each method, the data can be downloaded to the UCI database (http://archive.ics.uci.edu/ml/index.php).Supervised learning: Train to get an optimal model (which is a set of functions, the optimal one is best under a certain evaluation criterion) through an existing training sample (know

Coursera Andrew Ng Machine learning is really too hot, recently had time to spend 20 days (3 hours a day or so) finally finished learning all the courses, summarized as follows:(1) Suitable for getting started, speaking the comparative basis, Andrew speaks great;(2) The exercise is relatively easy, but to carefully consider each English word, or easy to make mist

Recently did a lot of Kaggle machine learning contest, summed up in addition to an experience: Do feature enginering can go to the former 20, if you want to enter the first 10, then need ensemble method support, So recently, we have developed a thorough understanding of the following combinations of methods. Through learning to find the combination method is real

1 reviews1.1 Supervised learningDefinition: Machine learning algorithm for given correct answerClassification:(1) Regression algorithm: Predicting the output of successive values, such as the prediction of house prices(2) Classification algorithm: The output of discrete values, such as determining whether the disease is a certain type of cancer1.2 Non-supervised learningDefinition: The relationship between

Problem Description:Given linear data set: t={(X1,y1), (x2,y2),..., (Xn,yn)}, presence of hyper plane S: $w \cdot x+b=0$$ \left\{\begin{matrix} w\cdot x+b>0,y=+1\\ W\cdot x+bLearning Strategies:The distance to define the point x0 to the super plane S is:$\frac{1}{\left \| w \right \|} \left | W \cdot x +b \right |$For mis-categorized data $ (X_{i},y_{i}) $, $-y_{i} (w \cdot x_{i}+b) >0$So when the distance between the x_{i} to the plane S is mistakenly classified: $-\frac{1}{\left \| w \right \|

* *.Second, installation Scikit-learnExecute command:Conda Install Scikit-learnSecond, installation KrasExecute command:Conda Install KerasThe required tensorflow is automatically installation during installation of the Keras process.At this point, deep learning, machine learning development environment has been installed, you can commandSpyderOrJupyter Notebook

First, Introduction1. Concept : The field of study that gives computers the ability to learn without being explicitly programmed. --an older, informal definition by Arthur Samuel (for tasks that cannot be programmed directly to enable the machine to learn) "A computer program was said to learn from experience E with respect to some class of tasks T and performance measure P, if Its performance on tasks in T, as measured by P, improves wit

Liblinear instead of LIBSVM 2.Liblinear use, Java version Http://www.cnblogs.com/tec-vegetables/p/4046437.html 3.Liblinear use, official translation. http://blog.csdn.net/zouxy09/article/details/10947323/ http://blog.csdn.net/zouxy09/article/details/10947411 4. Here is an article, write good. Transferred from: http://blog.chinaunix.net/uid-20761674-id-4840097.html For the past more than 10 years, support vector machines (SVM machines) have been the most influential algorithms in

Recommended book:Data mining: Practical machine learningData mining: Concepts and Techniques Han Jiawei; Read + reference articles later;Machine learning Combat (python);Machine learning Practical Case Analysis (r language);Neural networks and

Environment construction process is very troublesome ... But finally is ready, first give some of the process of reference to the more important information (find Microsoft's machine learning materials is a personal experience, without any reference):1. If the online various numpy, scipy and so on package installation tutorial trouble, go directly to: Microsoft Machine