Data Science Career

As a VI student, you will be given access to online learning materials in our e-learning portal.

To get you ready for learning, this essential pre-work will familiarise you with the basics of the key concepts and tools we will be using throughout the course.

Although you will learn these topics remotely before you arrive in class, you won’t be far away from the resources of the VI community. Make use of our Telegram channel to leverage connections with students, alumni, instructors and experts. At the end of your pre-work, you’ll be ready for the fast pace on campus!

After the course, you can choose to participate in follow-up sessions with your instructor, either in a group and/or individually, included as part of the course fee.

Tutorials:

• Introduction to Data Science
  • What is Data Science?
  • Data Science Life Cycle
  • What is Python and Why Learn It?
  • Introduction to Jupyter Notebook
• Introduction to Python Fundamentals  
  • Introduction to Data Types
  • Python Variables (In-built Functions)
  • Arithmetic, Relational and Logical Operators
  • Python Datatypes (String, Lists, Tuples, Dictionaries)
  • None and Casting
• Introduce OpenAI for generating Python code, debugging, and validating formulas

Learning Objectives:

• Describe data science and its life cycle, Python as a programming language and how to operate it on Jupyter Notebook editor
• Utilize Python to perform arithmetic, relation and logical operators with different data types and functions

• String Methods  
  • String Indexing
  • String Concatenation
  • String Formatting
  • List Slicing
• Python Iterations, Control Flow, and Functions (if…else statements, for and while loops)

Learning Objectives:

• Utilize different methods of data slicing and manipulation on textual data, iterations for repetitive tasks and user defined functions for control and logic flow

• Numpy 
  • Introduction to NumPy
  • Properties of Ndarray 
  • Basic Operations of Ndarray Object (Arithmetic Operations) 
  • Indexing and Iterations
  • Importing Packages
• Pandas 
  • Introduction to Pandas
  • Basic Operations of Series (Arithmetic Operations, Evaluating Values)
  • Basic Operations of Dataframes (Mathematical Operations)
  • Importing Files into Dataframes
  • Joins in Pandas (Merge & Concat)

Learning Objectives:

• Interpret Numpy arrays and use Numpy statistical module in Python for data indexing and iterations
• Perform essential operations of Pandas module to extract and manipulate data to prepare for statistical process and machine learning

• Data Visualisation- Matplotlib & Seaborn
  • Introduction to Matplotlib
  • Barplot, Histogram/Density Plot, Line Chart, Scatter Plot, Boxplot, Heatmap
  • Graph Parameters (Changing Size, Color, Style Markers, Titles, Legends and Label Orientation) in Matplotlib
• Data Cleaning and Exploratory Data Analysis 
  • Introduction to Data Cleaning
  • Common steps in Data Cleaning
  • Exploratory Data Analysis (Filtering and Sorting, Column Manipulation, Group By/Aggregate Functions, Handling Missing Data, using Functions)

Learning Objectives:

• Build visualizations and graphs using Matplotlib and Seaborn to interpret data and gather insights to support business decisions
• Use advanced techniques to clean data to conduct exploratory analysis for better understanding of the data

• Linear Regression
  • Modeling and Predictions
  • Introduction to Linear Regression
  • Introduction to Scikit-Learn Package (Fitting the Data, Evaluation of Model and Comparing Models)
  • Introduction to Statsmodels
• Feature Scaling
  • Standardisation
  • MinMax

Learning Objectives:

• Apply linear regression machine learning model to predict results of a continuous data type and evaluate its model performance
• Use advanced techniques in standardization and normalisation of the datasets to obtain better performance in machine learning models

• K-nearest neighbors 
  • Introduction to Classification
  • Introduction to KNN
  • Advantages and Drawbacks of KNN
  • Training KNN using Scikit-Learn 
• Logistic Regression 
  • Binary Class
  • Probability Estimation Dilemma
  • Odds Ratio
  • Log Odds
• Decision Trees and Random Forest Classification
  • Algorithm Walk-Through
  • Advantages and Drawbacks of Decision Trees
  • Training Decision Tree using scikit-learn 
  • Training Random Forest Scikit-Learn

Learning Objectives:

• Apply different algorithms to predict binary or multiclass outcomes and evaluate the model performances among the algorithms

• Prologue to GridSearch 
  • Introduction to GridSearch
  • Review of Initial EDA Strategies
  • Implement Changes and Updates to KNN Model using GridSearch
  • Find Optimal Hyperparameters of a model
  • Apply GridSearch to Classification Model
• Sklearn Pipelines
  • Inspecting Pipelines
  • Pipelines with GridSearch
  • Cross Validation

Learning Objectives:

• Create and automate hypertuning of machine learning models to obtain optimal machine learning parameters for good machine learning results

Participants will be required to attend an assessment after the last lesson as part of the funding requirements.