Three-Project Series

End-to-End Time Series Forecasting with Deep Learning you own this product

prerequisites
intermediate Python • basic SQL • intermediate data science • basics of deep learning • basics of data visualization
skills learned
build time series forecasting models • enhance the models with deep learning • serve predictions with a REST API using FastAPI
Jiahao Weng
3 weeks · 6-8 hours per week average · INTERMEDIATE
includes 3 liveProjects
liveProject $49.99 $59.99 self-paced learning

In this liveProject series, you’ll learn how to deliver an end-to-end machine learning application for time series forecasting. Taking on the role of a data scientist at a huge retailer, you’ll build time series forecasting models to anticipate the future so your bosses can make better decisions. You’ll go all the way through from creating your baseline testing models, to enhancing with deep learning, to even deploying your model as an easy-to-use usable application. The skills you learn are perfect for solving some of the most complex problems of data science, and are in high demand across the industry.

These projects are designed for learning purposes and are not complete, production-ready applications or solutions.

here's what's included

Project 1 Prepare Time Series Data

In this liveProject, you’ll create baseline models with Naive and sNaive methods for time series forecasts that you can use as a point of comparison for other models. Taking on the role of a data scientist for a large retail company, you’ll go hands-on to prepare sales data, create the baseline models, and optimize a Prophet model to compare against your baseline.

$29.99 $19.99
add to cart
Project 2 Forecast using Deep Learning

In this liveProject, you’ll use deep learning to implement powerful time series forecasting models that can beat the performances of previous models. You’ll work with the Python package “PyTorch Forecasting” and the deep learning models LSTM and N-BEATS. You’ll also get experience with key techniques of cross learning, ensembling, and hyperparameter tuning.

$29.99 $19.99
add to cart
Project 3 Deploy a Forecasting Model

In this liveProject, you’ll architect a solution to serve predictions from time series forecasting models over a REST API. Once you’ve architected your solution from a high-level perspective, you’ll monitor and assess the performance of the model and potentially undertake retraining to improve accuracy.

$29.99 $19.99
add to cart

books resources

When you start each of the projects in this series, you'll get full access to the following books for 90 days.

project author

Jiahao Weng
Jiahao Weng is a machine learning practitioner and a senior data scientist at a multinational company where he delivers projects ranging from proof-of-concept to production machine learning systems. As a freelance writer, he also contributes data science Medium articles to share his knowledge with the community.

Prerequisites

This liveProject series is for intermediate data scientists interested in tackling their first end-to-end deep learning project. To begin this liveProject, you will need to be familiar with the following:


TOOLS
  • Intermediate Python
  • Basic Google Colab
  • Basic Jupyter Notebook
  • Basic SQL
TECHNIQUES
  • Intermediate data science
  • Basics of data visualization
  • Basics of deep learning

you will learn

In this liveProject series, you’ll tackle different areas of forecasting and model building. The skills you learn are the same kind used to solve complex problems by forecasters and data scientists in the industry.


  • Process and clean time series data with outlier detection and linear interpolation
  • Analyze and visualize time series with Seaborn and Plotly charts
  • Determine seasonality with periodogram
  • Train-test split time series data using nested walk-forward validation
  • Establish baseline models with Naive and sNaive methods
  • Set up and optimize a Prophet model as an alternative model for comparison
  • Evaluate various models using weighted MAE and MAPE
  • Perform forecasting with LSTM and N-BEATS models using PyTorch Forecasting and PyTorch Lightning
  • Understand the concept of cross learning
  • Enhance model performance with ensembling technique
  • Optimize model with Bayesian optimization using Optuna
  • Implement callbacks for customized model training
  • Evaluate model training with TensorBoard
  • Read and write data to a PostgreSQL database with SQLAlchemy and pandas
  • Structure a code repository for easier code maintenance
  • Serve predictions with a REST API using FastAPI
  • Validate the API query parameters by using Python’s type annotations and pydantic
  • Track our model’s performance with MLflow
  • Retrain, compare, and update models on a periodic basis

features

Self-paced
You choose the schedule and decide how much time to invest as you build your project.
Project roadmap
Each project is divided into several achievable steps.
Get Help
While within the liveProject platform, get help from other participants and our expert mentors.
Compare with others
For each step, compare your deliverable to the solutions by the author and other participants.
book resources
Get full access to select books for 90 days. Permanent access to excerpts from Manning products are also included, as well as references to other resources.
RECENTLY VIEWED