Python for Data Science Training Overview
This Data Science with Python training teaches attendees how to use the Python Programming language for modern data science. Whether your team needs Python training for data analytics, machine learning, or data engineering, this course provides a solid introduction to Python.
Location and Pricing
Accelebrate offers instructor-led enterprise training for groups of 3 or more online or at your site. Most Accelebrate classes can be flexibly scheduled for your group, including delivery in half-day segments across a week or set of weeks. To receive a customized proposal and price quote for private corporate training on-site or online, please contact us.
In addition, we offer some courses as live, instructor-led online classes for individuals.
Objectives
- Understand the data processing phases
- Master Jupyter Notebooks
- Use NumPy and pandas
- Understand data repairing and normalization techniques
- Work with data visualization and Exploratory Data Analysis (EDA)
- Discover the Machine Learning algorithms of the scikit-learn Python library
Prerequisites
Participants must have modern programming or scripting experience and be familiar with core statistical concepts (i.e., variance, correlation, etc.).
Outline
Expand All | Collapse All
Python for Data Science
- Python Data Science-Centric Libraries
- SciPy
- NumPy
- pandas
- Scikit-learn
- Matplotlib
- Seaborn
- Python Dev Tools and REPLs
- IPython
- Jupyter Notebooks
- Anaconda
Defining Data Science
- What is Data Science?
- Data Science, Machine Learning, AI?
- The Data-Related Roles
- The Data Science Ecosystem
- Tools of the Trade
- Who is a Data Scientist?
- Data Scientists at Work
- Examples of Data Science Projects
- An Example of a Data Product
- Applied Data Science at Google
- Data Science Gotchas
Data Processing Phases
- Typical Data Processing Pipeline
- Data Discovery Phase
- Data Harvesting Phase
- Data Priming Phase
- Exploratory Data Analysis
- Model Planning Phase
- Model Building Phase
- Communicating the Results
- Production Roll-out
- Data Logistics and Data Governance
- Data Processing Workflow Engines
- Apache Airflow
- Data Lineage and Provenance
- Apache NiFi
Practical Introduction to NumPy- The First Take on NumPy Arrays
- The ndarray Data Structure
- Getting Help
- Understanding Axes
- Indexing Elements in a NumPy Array
- Understanding Types
- Re-Shaping
- Commonly Used Array Metrics
- Commonly Used Aggregate Functions
- Sorting Arrays
- Vectorization
- Vectorization Visually
- Broadcasting
- Broadcasting Visually
- Filtering
- Array Arithmetic Operations
- Reductions: Finding the Sum of Elements by Axis
- Array Slicing
- 2-D Array Slicing
- Slicing and Stepping Through
- The Linear Algebra Functions
Practical Introduction to pandas
- What is pandas?
- The Series Object
- Accessing Values and Indexes in Series
- Setting Up Your Own Index
- Using the Series Index as a Lookup Key
- Can I Pack a Python Dictionary into a Series?
- The DataFrame Object
- The DataFrame's Value Proposition
- Creating a pandas DataFrame
- Getting DataFrame Metrics
- Accessing DataFrame Columns
- Accessing DataFrame Rows
- Accessing DataFrame Cells
- Using iloc
- Using loc
- Examples of Using loc
- DataFrames are Mutable via Object Reference!
- The Axes
- Deleting Rows and Columns
- Adding a New Column to a DataFrame
- Appending / Concatenating DataFrame and Series Objects
- Example of Appending / Concatenating DataFrames
- Re-indexing Series and DataFrames
- Getting Descriptive Statistics of DataFrame Columns
- Navigating Rows and Columns For Data Reduction
- Getting Descriptive Statistics of DataFrames
- Applying a Function
- Sorting DataFrames
- Reading From CSV Files
- Writing to the System Clipboard
- Writing to a CSV File
- Fine-Tuning the Column Data Types
- Changing the Type of a Column
- What May Go Wrong with Type Conversion
Repairing and Normalizing Data
- Repairing and Normalizing Data
- Dealing with the Missing Data
- Sample Data Set
- Getting Info on Null Data
- Dropping a Column
- Interpolating Missing Data in pandas
- Replacing the Missing Values with the Mean Value
- Scaling (Normalizing) the Data
- Data Preprocessing with scikit-learn
- Scaling with the scale() Function
- The MinMaxScaler Object
Data Visualization in Python
- Data Visualization
- Data Visualization in Python
- Matplotlib
- Getting Started with matplotlib
- The matplotlib.pyplot.plot() Function
- The matplotlib.pyplot.bar() Function
- The matplotlib.pyplot.pie () Function
- The matplotlib.pyplot.subplot() Function
- A Subplot Example
- Figures
- Saving Figures to a File
- Seaborn
- Getting Started with seaborn
- Histograms and KDE
- Plotting Bivariate Distributions
- Scatter plots in seaborn
- Pair plots in seaborn
- Heatmaps
- A Seaborn Scatterplot with Varying Point Sizes and Hues
- ggplot
Descriptive Statistics Computing in Python
- Descriptive Statistics
- Non-uniformity of a Probability Distribution
- Using NumPy for Calculating Descriptive Statistics Measures
- Finding Min and Max in NumPy
- Using pandas for Calculating Descriptive Statistics Measures
- Correlation
- Regression and Correlation
- Covariance
- Getting Pairwise Correlation and Covariance Measures
- Finding Min and Max in pandas DataFrame
Data Science and ML Terminology and Concepts
- In-Class Discussion
- Types of Machine Learning
- Terminology: Features and Observations
- Representing Observations
- Terminology: Labels
- Terminology: Continuous and Categorical Features
- Continuous Features
- Categorical Features
- Features Types Visually
- Common Distance Metrics
- The Euclidean Distance
- What is a Model
- Data Scientist's Typical Activities
- The Typical Machine Learning Process
- Supervised vs Unsupervised Machine Learning
- Supervised Machine Learning Algorithms
- Unsupervised Machine Learning Algorithms
- Which Algorithm to Choose?
- The scikit-learn Package
- scikit-learn Estimators, Models, and Predictors
- Model Evaluation
- The Error Rate
- Confusion Matrix
- The Binary Classification Confusion Matrix
- Multi-class Classification Confusion Matrix Example
- Feature Engineering
- Scaling of the Features
- Feature Blending (Creating Synthetic Features)
- The 'One-Hot' Encoding Scheme
- Example of a 'One-Hot' Encoding Scheme
- Bias-Variance (Underfitting vs. Overfitting) Trade-off
- The Modeling Error Factors
- One Way to Visualize Bias and Variance
- Underfitting vs. Overfitting Visualization
- Balancing Off the Bias-Variance Ratio
- Regularization in scikit-learn
- Regularization, Take Two
- Dimensionality Reduction
- The Advantages of Dimensionality Reduction
- A Better Algorithm or More Data?
- The LIBSVM format
- Data Splitting into Training and Test Datasets
- ML Model Tuning Visually
- Data Splitting in scikit-learn
- Cross-Validation Technique
Classification Algorithms (Supervised Machine Learning)
- Classification (Supervised ML) Use Cases
- Classifying with k-Nearest Neighbors
- k-Nearest Neighbors Algorithm Visually
- Decision Trees
- Decision Tree Terminology
- Decision Tree Classification in the Context of Information Theory
- Using Decision Trees
- Properties of the Decision Tree Algorithm
- The Simplified Decision Tree Algorithm
- Random Forest
- Properties of the Random Forest Algorithm
- Support Vector Machines (SVMs)
- SVM Classification Visually
- Properties of SVMs
- Dealing with Non-Linear Class Boundaries
- Logistic Regression (Logit)
- The Sigmoid Function
- Logistic Regression Classification Example
- Logistic Regression's Problem Domain
- Naive Bayes Classifier (SL)
- Naive Bayesian Probabilistic Model in a Nutshell
- Bayes Formula
- Document Classification with Naive Bayes
Linear Regression
- Regression Analysis
- Regression vs. Correlation
- Regression vs. Classification
- Simple Linear Regression Model
- Linear Regression Illustration
- Least-Squares Method (LSM)
- Gradient Descent Optimization
- Multiple Regression Analysis
- Evaluating Regression Model Accuracy
- The R2 Model Score
- The MSE and RMSD Model Scores
- Gradient Boosting
- XGBoost
Unsupervised Machine Learning Algorithms
- PCA
- PCA and Data Variance
- PCA Properties
- Importance of Feature Scaling Visually
- Unsupervised Learning Type: Clustering
- Clustering vs. Classification
- Clustering Examples
- k-means Clustering in a Nutshell
- k-means Characteristics
- Global vs. Local Minimum Explained
Quick Introduction to Python for Data Engineers (Optional)
- What is Python?
- Additional Documentation
- Which version of Python am I running?
- Python Dev Tools and REPLs
- IPython
- Jupyter
- Jupyter Operation Modes
- Jupyter Common Commands
- Anaconda
- Python Variables and Basic Syntax
- Variable Scopes
- PEP8
- The Python Programs
- Getting Help
- Variable Types
- Assigning Multiple Values to Multiple Variables
- Null (None)
- Strings
- Finding the Index of a Substring
- String Splitting
- Triple-Delimited String Literals
- Raw String Literals
- String Formatting and Interpolation
- Boolean
- Boolean Operators
- Numbers
- Looking Up the Runtime Type of a Variable
- Divisions
- Assignment-with-Operation
- Relational Operators
- The if-elif-else Triad
- An if-elif-else Example
- Conditional Expressions (a.k.a. Ternary Operator)
- The While-Break-Continue Triad
- The for Loop
- try-except-finally
- Lists
- Main List Methods
- Dictionaries
- Working with Dictionaries
- Sets
- Common Set Operations
- Set Operations Examples
- Finding Unique Elements in a List
- Enumerate
- Tuples
- Unpacking Tuples
- Functions
- Dealing with Arbitrary Number of Parameters
- Keyword Function Parameters
- The range Object
- Random Numbers
- Python Modules
- Importing Modules
- Installing Modules
- Listing Methods in a Module
- Creating Your Own Modules
- Creating a Runnable Application
- List Comprehension
- Zipping Lists
- Working with Files
- Reading and Writing Files
- Reading Command-Line Parameters
- Accessing Environment Variables
- What is Functional Programming (FP)?
- Terminology: Higher-Order Functions
- Lambda Functions in Python
- Example: Lambdas in the Sorted Function
- Other Examples of Using Lambdas
- Regular Expressions
- Using Regular Expressions Examples
- Python Data Science-Centric Libraries
Conclusion
Training Materials
All Python Data Science training attendees receive comprehensive courseware.
Software Requirements
- Anaconda Python 3.6 or later
- Spyder IDE and Jupyter notebook (Comes with Anaconda)
