Duration
1 day
What You will Learn
- Data cleaning or cleansing as a pre-processing step towards making the data more consistent and high quality before training predictive models.
- Improve your understanding of descriptive statistics and apply them over a dataset.
- Learn how to deal with missing data and outliers to resolve data inconsistencies.
- Explore various visualization techniques for bivariate and multivariate analysis.
- Learn multidimensional analysis and reduction techniques.
Course Outline
Introduction to the Data Science in Python
- What is Data Science?
- Introduction to the Python Data Science Tool
- Introduction to the Python Data Science Environment
- Introduction to Python Pre-Requisites for Data Science
- Different Types of Data Used in Statistical & ML Analysis
- Different Types of Data Used Programmatically
- Python Data Science Packages To Be Used
Exploratory Data Analysis
- What is Exploratory Data Analysis?
- Exploratory Data Analysis vs Data Analysis
- Exploratory Data Analysis Techniques
- Exploratory Data Analysis Process
- Types of Exploratory Data Analysis
Introduction to Numpy
- Numpy: Introduction
- Create Numpy Arrays
- Numpy Operations
- Matrix Arithmetic and Linear Systems
- Numpy for Basic Vector Arithmetric
- Numpy for Basic Matrix Arithmetic
- Broadcasting with Numpy
- Solve Equations with Numpy
- Numpy for Statistical Operation
Introduction to Pandas
- Data Structures in Python
- Read in Data
- Read in CSV Data Using Pandas
- Read in Excel Data Using Pandas
Introduction to Data Visualizations
- What is Data Visualization?
- Some Theoretical Principles Behind Data Visualization
- Histograms-Visualize the Distribution of Continuous Numerical Variables
- Boxplots-Visualize the Distribution of Continuous Numerical Variables
- Scatter Plot-Visualize the Relationship Between 2 Continuous Variables
- Barplot
- Pie Chart
- Line Chart
Data cleaning and Feature Engineering
Introduction
- Types of Variables
- Variable Characteristics
- Missing Data Imputation
- Multivariate Imputation
- Categorical Encoding
- Variable transformation
- Discretisation
- Outlier Engineering
- Feature Scaling
- Mixed Variables
- Engineering Date-Time
Programme scope, delivery arrangements and suitability are confirmed during enquiry.