Readme

I will be regularly updating this repository to include more tutorials on Python with streamlit for machine learning and machine vision applications. Below is an overview of available demos. Check the corresponding subfolders for more details, code, and other explanation.

If you have no experience with Python programming, then you need to start Here, there is a list of resources for learning Python and relevant tools, and also a notebook with some Python code covering the basic concepts such as variables, loops, etc…

(1) Simple Linear Model

A tutorial on linear regression using Python (basics), will teach you more python skills, and the basics of Linear Models. Also, we will use streamlit to build an interactive application to all interaction with the underlying models. More details can be found here Python Intro. By the end of this tutorial, you should be able to have an application like the one below to perform simple predictions

(2) Logistic Regression

Logistic Regression is a Supervised Machine Learning classification algorithm. We used it to predict the probability that an event will occur. The response variable (what we try to predict), is a binary variable. In other words it has two classes (i.e.~pass/ fail). Often represented as 0 and 1 respectively. The Logistic Regression Model predicts (P(y=1, | X)), which reads as the probability that (y) will happen given (X), where (X) is the set of features (see lecture notes). Code can be found Here

(3) Face Recognition using SVM

This is a quick demo using SVM to recognise faces from images. It also uses streamlit for creating the interactive front-end features. More details at the SVMFR repository.

(4) Symbols Classification using Random Forests

The tutorial will focus on classification of symbols, more specifically:

• You will learn how to load, view and classify instances representing 2D images
• Create and evaluate A Random Forest Model Classifier (Bagging ensemble-based method) using sklearn.ensemble.RandomForestClassifie library. Code is available here as a Jupyter Notebook
• Make predictions and visualise the results
• Use streamlit to create an interactive front end to make predictions and view results. The python file that handles the streamlit part is Here. By the end of the tutorial, you will be able to produce something similar to this Demo. You should be able to produce a solution like the one below by the end of this tutorial using Python, streamlit and relevant resources

Comments / Questions

You can reach me at my staff page or on linkedin