2018 Most popular Python machine learning Library Introduction

python is an object-oriented, interpretive computer programming language with a rich and powerful library, coupled with its simplicity, ease of learning, speed, open source free, portability, extensibility, and object-oriented features,python Become the most popular programming language of the 2017!

AI is one of the most popular topics, machine learning technology is an artificial intelligence to achieve the necessary skills, Python programming language contains the most useful machine learning tools and libraries, the following is the Python development engineers must know the top ten machine learning library!

One, Scikit-learn

II, statsmodels

  statsmodels is another powerful library focused on statistical models, primarily for predictive and exploratory analysis, fitting linear models, For statistical analysis or predictive modelling, it is very appropriate to use  statsmodels.

Third, PYMC

PYMC is a tool for "Bayesian curve", which contains Bayesian model, statistical distribution and model convergence diagnostic tools, but also contains some hierarchical models.

Iv. Gensim

  gensim is called the "People's Theme Modeling tool", which focuses on Dirichlet division and variants, which support natural language processing, It is easier to combine NLP and other machine learning algorithms, and it also cites the Word2vec of Google's text notation based on recurrent neural networks.

Five, Orange

Orange is a library with a graphical user interface that is quite complete in terms of classification, aggregation, and feature selection methods, as well as cross-validation methods.

Liu, PYMVPA

PYMVPA is a statistical learning library that includes cross-validation and diagnostic tools, but not scikit-learn comprehensive.

Vii. Theano

  theano is the most mature deep learning library that provides a nice data structure representing the layers of a neural network, which is efficient for linear algebra, Similar to the NumPy array, many Theano-based libraries are leveraging their data structures, which also support out-of-the-box GPU programming.

Eight, Pylearn

Pylearn is a Theano-based library that introduces modularity and configuration to Theano, which can be used to create neural networks through different configuration files.

IX, Hebel

  hebel is a neural network library with GPU support that can determine the properties of a neural network through YAML files. Provides a way to separate the Divine Network and code-friendly, and run the model quickly, it is written in pure Python, is a very friendly library, but because of the development soon, on the depth and the vast, there is some lack!

ten, Neurolab

  neurolab is an API-friendly neural network library that contains different variants of the recursive neural network implementation, if RNN is used, This library is one of the best choices in the same API.

The above is the Python development engineer must know ten machine learning Library, in addition, there are overfeat, Nolearn and decaf, such as machine learning Library, in this will not be introduced, interested in in-depth study!

2018 Most popular Python machine learning Library Introduction