This course explores the essential steps for preparing data for machine learning, focusing specifically on financial time series data. From feature engineering to scaling and train-test splitting, you will learn to apply best practices in preprocessing data to pave the way for successful model training and evaluation.
Overview
Syllabus
- Unit 1: Feature Engineering for ML
- Modify Tesla Stock Data Columns
- Debug the Tesla Stock Code
- Creating and Inspecting New Financial Features
- Create New Features for Tesla Stock Data
- Tesla Stock Feature Engineering
- Unit 2: Scaling Features with StandardScaler
- Scaling a Single Feature with StandardScaler
- Identify and Fix the Code
- Scaling Financial Features with StandardScaler
- Implement Feature Scaling Using StandardScaler
- Final Data Scaling Implementation
- Unit 3: Splitting Dataset into Training and Testing Set
- Adjust the Dataset Split Ratio
- Fix the Dataset Split
- Fill in the Blanks: Splitting and Scaling Data
- Splitting the Dataset into Training and Testing Sets
- Preprocess and Split Tesla Stock Data
- Unit 4: Addressing Data Leakage in Time Series
- Adjusting TimeSeriesSplit to 5 Splits
- Fixing Time Series Data Split
- Ensure Proper Scaling in Time Series Splitting
- Feature Scaling and Time Series Split
- Addressing Data Leakage in Time Series
- Unit 5: Creating Lag Features for Time Series Prediction
- Creating Lag Features for Two Days
- Fix the Lag Feature Code
- Adding Lag Features and Handling NaN Values
- Creating and Using Lag Features for Stock Price Prediction