Course Overview
This course aims to be a comprehensive course in using the power of Python to read data, prepare the data for analysis, create exploratory visualizations, and train machine learning algorithms. Students will also be sufficiently equipped to start their journey towards becoming full-fledged Data Scientists.
Course Objectives
Extract, transform and visualize the data using Python
Implement advanced data transformation.
Understand the statistical background of Machine learning algorithms.
Implement end-to-end AI starting with the data, train the machine learning algorithm and finally publish it as AI
Duration: 5 Days
Outlines
--Day 1
Chapter 1 – Python Basics (9:00 AM-12:00 PM)
Download and Install Python
How to run a Python script
Python data types o Numbers o Strings o Dates o Lists o Tuples o Dictionaries
Conditions
Lunch (12:00 PM-1:00 PM)
(1:00 PM-5:00 PM)
Loops
Functions
Modules
Reading a text file
Writing to a text file
Reading an Excel file
Overview of Jupiter notebook
The Basics of NumPy Arrays
--Day 2
Chapter 2– Data Manipulation and Visulization with Pandas and Matplotlib
(9:00 AM-12:00 PM)
Installing Pandas and Matplotlib
DataFrame exercise
Data Indexing and Selection
Operation on Data in Pandas o Handling Missing Data o Combining DataSets: Concat and Append o Combining DataSets: Merge and Join o Aggregation and Grouping
Lunch (12:00 PM-1:00 PM)
Visualization with Matplotlib (1:00 PM-5:00 PM) o General Matplotlib Tips o Simple Line Plot o Simple Scatter Plot o Histograms o Customizing Plot Legends o Customizing Colour bars o Multiple Subplots o Text and Annotation
--Day 3
Chapter 3 – Machine Learning
(9:00 AM-12:00 PM)
What is Machine Learning?
Data Types in machine learning
Data considerations
Statistics fundamentals o Mean o Median o Mode o Standard Deviation o Variance
Feature Engineering
Lunch (12:00 PM-1:00 PM)
(1:00 PM-5:00 PM)
Introducing Scikit-Learn o Data Transformation (remove duplication and replace null values) o Data Scaling o Train o Evaluate o Predict
Overview of Overfit and bad fit
--Day 4
Chapter 4 – Types of Algorithms
(9:00 AM-12:00 PM)
Regression o How does it work and its use cases o Linear regression o Polynomial Regression
Train a regression model
Evaluate the regression model o R-Squared o Mean Absolute Error
Classification o How does it work and its use cases
Train a classification model
Evaluate the classification model o Accuracy o Precision o Recall o F-score
Lunch (12:00 PM-1:00 PM)
(1:00 PM-5:00 PM)
Clustering o How does it work and its use cases
Fit clustering algorithm model
Evaluate the clustering model o Silhouette Score to choose the best number of clusters
Three projects to do
--Day 5
Chapter 5- Publish Machine Learning Model as API
(9:00 AM-12:00 PM)
Overview of Flask web server
Loading HTML page
Post request implementation
Loading the Machine learning model
Perform prediction
Project to do
Lunch (12:00 PM-1:00 PM)
(1:00 PM-5:00 PM)
Anomaly Detection example
Introduction to Convolutional Neural Network for image processing (CNN)
TensorFlow Project for image classification
Q&A
Programme scope, delivery arrangements and suitability are confirmed during enquiry.