Description
Description
The “Data Science Projects with Python“ course aims to familiarize you with the Python environment tailored for data science tasks. It serves as a stepping stone in your quest to become proficient in various machine learning topics. These acquired skills will enable you to develop advanced predictive models, aligning with industry standards and delivering substantial value to businesses across diverse sectors.
“Data Science Projects with Python“ is meticulously crafted to provide practical guidance on utilizing industry-standard data analysis and machine learning tools in Python, utilizing real-world datasets. This course empowers you to grasp how to leverage pandas and Matplotlib to meticulously examine datasets through summary statistics and visualizations, extracting the desired insights effectively.
Continuing your learning journey, you will delve into data preparation techniques and the application of machine learning algorithms, such as regularized logistic regression and random forest, utilizing the scikit-learn package. You will explore techniques for fine-tuning these algorithms to yield optimal predictions on new and unseen data instances.
Advancing through subsequent chapters, you will gain a comprehensive understanding of the inner workings and outputs of these algorithms. This knowledge will not only enhance your predictive modeling capabilities but also enable you to comprehend the rationale behind these predictions.
Upon completing this course, you will possess the skills necessary to proficiently employ various machine learning algorithms for in-depth data analysis and extract actionable insights from datasets with confidence.
Hands on Training
Data Science Projects with Python takes a case study approach to simulate the working conditions you will experience when applying data science and machine learning concepts. You will be presented with a problem and a data set and walked through the steps of defining an answerable question, deciding what analysis methods to use, and implementing all of this in Python to create a deliverable.
Training Objectives
- By the end of this course
- you will have the skills you need to confidently use various machine learning algorithms to perform detailed data analysis and extract meaningful insights from data.
Course Outline
- Lesson 1: Data Exploration and Cleaning<
- Python and the Anaconda Package Management System<br />
- Different Types of Data Science Problems<br />
- Loading the Case Study Data with Jupyter and pandas<br />
- Data Quality Assurance and Exploration<br />
- Exploring the Financial History Features in the Dataset<br />
- Activity 1: Exploring Remaining Financial Features in the Dataset<
- Lesson 2: Introduction to Scikit-Learn and Model Evaluation<
- Introduction<br />
- Model Performance Metrics for Binary Classification<br />
- Activity 2: Performing Logistic Regression with a New Feature and Creating a Precision-Recall Curve<
- Lesson 3: Details of Logistic Regression and Feature Exploration<
- Introduction<br />
- Examining the Relationships between Features and the Response<br />
- Univariate Feature Selection: What It Does and Doesn't Do<br />
- Building Cloud-Native Applications<br />
- Activity 3: Fitting a Logistic Regression Model and Directly Using the Coefficients<
- Lesson 4: The Bias-Variance Trade-off<
- Introduction<br />
- Estimating the Coefficients and Intercepts of Logistic Regression<br />
- Cross Validation: Choosing the Regularization Parameter and Other Hyperparameters<br />
- Activity 4: Cross-Validation and Feature Engineering with the Case Study Data<
- Lesson 5: Decision Trees and Random Forests<
- Introduction<br />
- Decision trees<br />
- Random Forests: Ensembles of Decision Trees<br />
- Activity 5: Cross-Validation Grid Search with Random Forest<
- Lesson 6: Imputation of Missing Data, Financial Analysis, and Delivery to Client<
- Introduction<br />
- Review of Modeling Results<br />
- Dealing with Missing Data: Imputation Strategies<br />
- Activity 6: Deriving Financial Insights<br />
- Final Thoughts on Delivering the Predictive Model to the Client
