Deep learning in production
An AI model delivers lasting value only when its operation is controlled. With MLOps, connect deployment, monitoring and maintenance to organise the transition to production. Strengthen your ability to coordinate data and operations teams around models monitored over time.
- Duration
- 3 days 21 hours
- Code
- FC-DLP Code
Presentation
In recent years, deep learning has progressed rapidly. Frameworks and libraries have been developed and regularly updated. Nevertheless, there remains a lack of established solutions to manage, deploy and scale models. The maturity of deep learning research infrastructure also remains limited.
This level 2 deep learning course provides the skills needed to design effective deep learning models. You will also learn to adapt existing models and implement them. Finally, you will explore ways to address recurring deep learning challenges, such as a lack of labelled data and hyperparameter search.
By the end of this 3-day course, you will be able to create a deep learning model and adapt it using more advanced settings to improve performance. Labs throughout the programme will enable you to test your skills in deploying deep learning models to production.
Objectives
After completing the deep learning in production course, you will achieve the following learning objectives:
- create deep learning models implemented with production-quality code;
- use different approaches to deploy models to production;
- make informed use of a specific deployment method.
Program
Training strategies for neural networks
- Automated hyperparameter search.
- Using transfer learning.
- Using semi-supervised learning.
- Applying domain adaptation.
Deploying deep learning models to production
- Reducing model size through pruning, quantisation and optimised matrix computation.
- Presentation and implementation of different deployment solutions.
- Introduction to Docker and FastAPI.
- Deploying a simplified model to the cloud.
Delivering an end-to-end team project (labs)
- Create an implementation project covering data retrieval through to production deployment. Participants choose the use case in groups of 2 to 3.
Audience
This course is intended for:
- data scientists wishing to specialise in deep learning;
- support teams responsible for data management or research and development teams;
- IT managers wishing to specialise in data;
- anyone wishing to retrain in Big Data and artificial intelligence technologies.
Prerequisites
The deep learning in production course requires the following prerequisite:
- completion of the deep learning fundamentals course.
Teaching and assessment methods
- Initial skills assessment
- Training materials provided to participants
- Continuous assessment throughout the course
- End-of-course feedback questionnaire
- Combination of theory and practical application
- Attendance records
- Post-course follow-up evaluation
- Practical exercises
- Case study
Course highlights
Theoretical and practical training delivered in person or online; learning materials; labs covering various real-world use cases and rapid skills development.
Dates and sessions
Choose the date and delivery format that suit you.
No upcoming sessions are currently available.
fr
en