Machine Learning With Python




Machine Learning with Python

26 March 2018


Added 01-Jan-1970

ython has become a powerful language of data science and is now commonly used as the leading programming language for predictive analytics and artificial intelligence. During this hands-on 3-day “Machine Learning with Python” training course, the attendees will learn to utilise Python’s libraries for predictive analytics on the real-world data. The course will explore practical applications of major scientific libraries such as NumPy, pandas, SciPy and matplotlib, as well as more specialised, machine learning oriented SciKit-Learn, Theano, TensorFlow, Keras and H2O for Python.

The course will provide theoretical and practical understanding of major machine learning techniques such as:

  • multiple linear regressions (including ridge and Lasso) and Generalized Linear Models e.g. binomial and multinomial logistic regressions as well as Poisson regressions,

  • classification methods e.g. naive Bayes, k-nearest neighbours, decision trees, random forests, support vector machines,

  • clustering and dimensionality reduction methods: k-means and principal component analysis,

  • introduction to neural networks and deep learning.

The structure of the course will include short theoretical lectures introducing each of the above machine learning methods and practical tutorials presenting applications of these techniques using Python language. Apart from the machine learning methods, the attendees will also learn other concepts associated with predictive analytics and machine learning:

  • feature extraction and engineering,
  • normalisation and standardisation methods,
  • model optimisation through parameter grid search,
  • model validation and accuracy metrics including confusion matrix, precision, recall, F1 score, ROC, log-loss, Gini, MSE, RMSE, R-squared etc.,
  • selected approaches for machine learning with Big Data using Python and its libraries.

The course will utilise Python 3.x (Anaconda distribution), with additional libraries e.g. SciKit-Learn, Theano, Keras, H2O, TensorFlow etc. The full list of packages will be confirmed with the attendees before the course.
During the tutorial on multi-core and GPU-accelerated machine learning in Python with H2O and TensorFlow, the attendees will be also provided with access to Mind Project computing cluster.