TensorFlow Developer Certificate - Time Series, Sequences, and Predictions
TensorFlow is a powerful open-source library for machine learning and numerical computation to develop and train deep learning models. This course will teach you how to analyze time series data and accurately forecast future events.
What you'll learn
Forecasting key business metrics accurately is crucial for making smart decisions and planning for the future. In this course, TensorFlow Developer Certificate - Time Series, Sequences, and Predictions, you’ll gain the ability to build predictive models for time series data using deep learning.
First, you’ll explore how to structure time series problems and preprocess data for modeling.
Next, you’ll discover how to choose and configure deep learning models like RNNs and CNNs for sequence forecasting tasks.
Finally, you’ll learn best practices for training, evaluating, and improving your forecasts over time.
When you’re finished with this course, you’ll have the skills and knowledge of TensorFlow and deep learning for time series needed to accurately forecast metrics like future sales, demand, stock prices, and more.
Table of contents
- Intended Outcome and Prerequisites 3m
- Defining Time Series and Sequences 7m
- Demo: Time Series Visualization 3m
- Data Preparation Techniques for Time Series Learning 4m
- Demo: Time Series Data Preparation 4m
- Addressing Sequence Bias in Time Series Analysis 3m
- Demo: Techniques for Identifying and Reducing Sequence Bias 4m
- Overview of Error Metrics 3m
- Understanding the Implications of Mean Absolute Error (MAE) 1m
- Techniques for Model Training and Hyperparameter Tuning 3m
- Demo: Implementing Model Training and Evaluation 4m
- Implementing Recurrent Neural Networks (RNNs) 2m
- Demo: Building an RNN for Time Series Prediction 3m
- Implementing Convolutional Neural Networks (CNNs) 2m
- Demo: Implementing a CNN for Time Series Prediction 3m