Overview

Machine Learning can be an incredibly beneficial tool to uncover hidden insights and predict future trends. This Machine Learning with Python course will give you all the tools you need to get started with machine learning. This Machine Learning with Python course dives into the basics of machine learning using Python, an approachable and well-known programming language. This hands-on workshop will take the audience from the basic steps of machine learning that is linear regression and logistic regression to more complex models such a decision trees and neural networks.

During the workshop, we will learn how to make supervised machine learning (classification and regression), when to use classification and when to use regression models and when to use classification models, all with practical and real datasets. Followed by unsupervised techniques such as hierarchal clustering and advanced AI neural network models.

Course Director:

Haytham Omar

Founder, Rescale Analytics
MIT Certified in Data Science & Big Data Analytics

Haytham Omar is a data scientist, consultant, and a trainer. His areas of expertise include Supply Chain, Business Intelligence, and Data Science. aytham developed the Inventorize package in R mainly used for supply chain analytics with more than 6000 Downloads so far. Omar is a certified instructor in Supply Chain Analytics and holds a Master’s Degree in Supply Chain Management along with various prestigious certifications from MIT and other reputable institutes.

His is highly skilled in following domains: R Coding languages, SQL,SAP ERP, Microsoft azure solutions, Orange for data mining, Any logic for simulation modelling, Advanced Excel, Tableau.

Course Outline

Module 01:

1. Understanding data types and structures
2. What is the role of statistics and probabilities in machine learning.
   • Preparing your data for machine learning.
   • Splitting your data inpython for training and testing.

In the first part as probably, it could be our first time with python, we will work with simple objects in python and learn how to import data frames, how to clean the data and remove empty values or impute missing values. Or maybe transformation of the data and removing outliers.

After that, we will understand why we need to split our data so we can have more confidence on the validity of our machine learning models. We will also know when to use seed randomizations.

Module 02:

• Working on a classification problem using several classification learning models.
• Working on a regression problem using several regression models and also non-linear models.
• Evaluation the results of the model.
• Using cross validation to verify the effectiveness of the model.

This part will focus on the different categories of machine learning and when do we use classification, regression or clustering.

We will start machine learning problems using classification techniques such as logistic regression and K-nearest neighbors, these techniques we will use on real time problems so audience can relation to problems from a business environment context. After that , we will start using regression models and decision tree models to predict a continuous problem.

Finally, we will test the accuracy of the models and the concepts of overfitting and why the models need to be flexible and not highly overfitted. We will also use cross validation out of sample techniques to be more certain about the effectiveness of the machine learning model being formulated.

Module 03:

• Introduction to unsupervised learning.
• What are the applications of unsupervised learning.
• Unsupervised hierarchal clustering.

Part three we will focus on unsupervised learning techniques and segmentation of the data based on their similarities. when do we apply unsupervised learning and use cases from business.

Finally, we will end the workshop with a simple example on neural networks , how propagation is being made? what is gradient descent? and finally why its so powerful and provides better results than normal models.

Finally, we will test the accuracy of the models and the concepts of overfitting, and why the models need to be flexible and not highly overfitted. We will also use cross-validation out of sample techniques to be more certain about the effectiveness of the machine learning model being formulated.

Workshop Instructions

• A good processing speed laptop.
• Anaconda python download and install before the workshop for python version 3.7. (Link: https://www.anaconda.com/distribution/)
• It is preferable that you have a basic understanding of python and you have attended our workshop statistical analysis with python. However, it is not mandatory as we take participants step by step from the beginning in the world of machine learning.

Who Should Attend

• Willing to learn machine learning algorithm with Python.
• Who has a deep interest in the practical application of machine learning to real world problems.
• Wishes to move beyond the basics and develop an understanding of the whole range of achine learning algorithms.
• With intermediate to advanced EXCEL knowledge who is unable to work with large datasets.
• Interested to present their findings in a professional and convincing manner.
• Who wishes to start or transit into a career as a data scientist.
• Who wants to apply machine learning to their domain.