"Data determines the upper limit of the machine learning algorithm and just approaching this limit as far as possible", this sentence explains the importance of good data in machine learning. Most take over direct data are not obvious, and not treated or that there is a lot of useless data, it requires some processing features, zoom feature and so on, to meet the requirements of the training data.
We will first contact Scikit-learn to use this machine learning library
Scikit-learn
- Python language learning machine tools
- Everyone applicable, can be reused in different contexts
- Built on NumPy, SciPy and matplotlib
- Open-source, commercially available - BSD License
- The current stable version 0.18
Since its 2007 release, scikit-learn has become the most to the power of machine learning Python library (library) a. scikit-learn support machine learning algorithms including classification, regression, clustering and dimension reduction. Some feature extraction (extracting features), data processing (processing data) and model assessment (evaluating models) module. As an extension Scipy library, scikit-learn is built on the basis of Python NumPy library and matplotlib on top. NumPy Python support allows efficient operation of a large number of multi-dimensional matrix of data, matplotlib provides a visual tool, SciPy model with many scientific computing. scikit-learn documentation complete, easy to use, rich API, making it popular in academia. Developers scikit-learn experiments with different algorithms, just a few lines of code can handle. scikit-learn many implementations include well-known machine learning algorithms, including LIBSVM and LIBLINEAR. Also it encapsulates other Python libraries, such as natural language processing library NLTK. In addition, scikit-learn it built a large data set, allowing developers to focus on algorithm design, saving time acquire and organize data sets.
Installation, then refer to the following steps: Create a