BDI Python Code Clinic – 28 July 11am (Microsoft Teams – the link will be provided week commencing 27 July)
More Machine Learning in Python with scikit-learn
To book, click here:
Following the success of the previous tutorial on Machine Learning in Python, we organise another session on this topic. We will be covering other aspects of ML this time, so this Code Clinic can be attended as a follow-up from the previous ML Python session, as well as a separate independent session – everyone is welcome to join.
We will continue using scikit-learn Python library (scikit-learn.org).
Last time we went through the simple example of solving a classification problem using ML, while this time we will pay attention to the regression problem.
Additionally, we will learn about the feature selection methods, which allow reducing the overfitting and improving the accuracy of the predictions, and will apply feature selection to our dataset. Then, we will evaluate the performance of different regression algorithms via such metrics as Mean Absolute Error, Mean Squared Error, and R^2.
At the end of the session, we may discuss the differences between supervised and unsupervised learning problems and look deeper at the available algorithms for each type of data.
The following tools will be used in this code clinic:
Python3 – www.python.org
Python SciPy libraries:
sklearn (shorten from scikit-learn)
Irina Chelysheva will be working in Spyder IDE – www.spyder-ide.org, but, please, feel free to use your favourite one