Custom Models, Layers, and Loss Functions with TensorFlow from Coursera

In this course, you will:

• Compare Functional and Sequential APIs, discover new models you can build with the Functional API, and build a model that produces multiple outputs including a Siamese network.

• Build custom loss functions (including the contrastive loss function used in a Siamese network) in order to measure how well a model is doing and help your neural network learn from training data.

• Build off of existing standard layers to create custom layers for your models, customize a network layer with a lambda layer, understand the differences between them, learn what makes up a custom layer, and explore activation functions.

• Build off of existing models to add custom functionality, learn how to define your own custom class instead of using the Functional or Sequential APIs, build models that can be inherited from the TensorFlow Model class, and build a residual network (ResNet) through defining a custom model class.

The DeepLearning.AI TensorFlow: Advanced Techniques Specialization introduces the features of TensorFlow that provide learners with more control over their model architecture and tools that help them create and train advanced ML models.

This Specialization is for early and mid-career software and machine learning engineers with a foundational understanding of TensorFlow who are looking to expand their knowledge and skill set by learning advanced TensorFlow features to build powerful models.

What's inside

Syllabus

Functional APIs

Compare how the Functional API differs from the Sequential API, and see how the Functional API gives you additional flexibility in designing models. Practice using the functional API and build a Siamese network!

Custom Loss Functions

Loss functions help measure how well a model is doing, and are used to help a neural network learn from the training data. Learn how to build custom loss functions, including the contrastive loss function that is used in a Siamese network.

Custom Layers

Custom layers give you the flexibility to implement models that use non-standard layers. Practice building off of existing standard layers to create custom layers for your models.

Custom Models

You can build off of existing models to add custom functionality. This week, extend the TensorFlow Model Class to build a ResNet model!

Bonus Content - Callbacks

Custom callbacks allow you to customize what your model outputs or how it behaves during training. This week, implement a custom callback to stop training once the callback detects overfitting.

Good to know

Know what's good

, what to watch for

, and possible dealbreakers

Explores Functional and Sequential APIs, which is standard in industry

Teaches how to build custom loss functions, including the contrastive loss function used in Siamese networks

Develops custom layers, a core skill for deep learning engineers

Taught by Laurence Moroney and Eddy Shyu, who are recognized for their work in deep learning

Examines how to build off of existing models to add custom functionality

Advises students to take other courses first as prerequisites

Reviews summary

Custom tensorflow models with code

learners say this is a well received, step-by-step advanced course about customizing models with TensorFlow. According to students, key features include classes on making custom loss functions, custom layers, and more. Additionally, students say the outstanding instructor, Laurence Moroney, is easy to follow. Note that in some instances, you may encounter some minor errors in assignments or quizzes.

Knowledgeable and engaging.

"Great instructor and very good topics."

"Laurence you are just awesome !!"

"Mr Laurence is the best he develops my programmation skills "

"What a mastered lecturer is Laurence! Good luck for him and Coursera team"

Well organized with clear explanations.

"Well and Clear Organised content and very friendly guided labs."

"Excellent course with great content. The videos explain everything needed to start coding."

"Excellent course content, videos, labs and assignment with clear instructions or guidelines."

"The instructors did e great job of explaining tensorflow advanced concepts as well as making the assignments simple to follow"

Create models, loss functions, and more.

"Great course to learn how to use Tensorflow to build flexible model"

"Advanced technique explained very easy way. Very informative course"

"This course is extrememly helpful to help me learn building custom model and layers"

"It really helped me a lot to formulate my custom loss function for my research works."

Activities

Be better prepared before your course. Deepen your understanding during and after it. Supplement your coursework and achieve mastery of the topics covered in Custom Models, Layers, and Loss Functions with TensorFlow with these activities:

Explore the Functional API in detail

Show steps

Deepen your understanding of the Functional API by following guided tutorials that cover advanced topics such as building custom models and layers.

Browse courses on Functional API

Show steps

Follow the TensorFlow tutorial on building custom models using the Functional API
Follow the TensorFlow tutorial on building custom layers using the Functional API
Experiment with building your own custom models and layers using the Functional API

Practice building custom loss functions

Show steps

Reinforce your understanding of custom loss functions by completing practice drills that involve implementing and using them in your own models.

Show steps

Implement the contrastive loss function used in a Siamese network
Implement a custom loss function for a specific problem in your field of interest
Compare the performance of different custom loss functions on a given dataset

Develop a custom model for a real-world problem

Show steps

Apply your skills by developing a custom model for a real-world problem that is relevant to your field of interest.

Browse courses on Custom Models

Show steps

Identify a real-world problem that can be solved using a custom TensorFlow model
Design and implement a custom model architecture using the TensorFlow Model class
Train and evaluate the custom model on a relevant dataset
Write a report or blog post describing the problem, the model, and the results

One other activity

Expand to see all activities and additional details

Show all four activities

Mentor junior TensorFlow developers

Show steps

Enhance your understanding of TensorFlow while helping others by mentoring junior developers in the TensorFlow community.

Browse courses on Mentoring

Show steps

Join a TensorFlow community forum or online mentoring platform
Offer your help to junior TensorFlow developers who are遇到困难
Provide guidance and support to help them overcome challenges and grow their TensorFlow skills

Career center

Learners who complete Custom Models, Layers, and Loss Functions with TensorFlow will develop knowledge and skills that may be useful to these careers:

Deep Learning Engineer

Deep Learning Engineers specialize in developing and deploying deep learning models for various applications such as image recognition, natural language processing, and speech recognition. This course may be helpful for aspiring Deep Learning Engineers who want to learn how to build custom models, layers, and loss functions in TensorFlow to create powerful deep learning models.

See salaries and explore the career path for Deep Learning Engineer

Machine Learning Engineer

Machine Learning Engineers build and maintain advanced ML models to power a wide range of products and services, and help businesses make more informed decisions. This course may be helpful for aspiring Machine Learning Engineers who wish to build advanced TensorFlow models and gain hands-on experience with custom models, layers, and loss functions.

See salaries and explore the career path for Machine Learning Engineer

Data Scientist

Data Scientists use their knowledge of statistics, programming, and machine learning to extract insights from data and help businesses make data-driven decisions. This course may be helpful for aspiring Data Scientists who want to learn how to build custom models, layers, and loss functions in TensorFlow to create advanced ML models for data analysis and prediction.

See salaries and explore the career path for Data Scientist

Software Engineer

Software Engineers design, develop, and maintain software systems. This course helps build a foundation in TensorFlow, which is a popular framework for building and training ML models. This knowledge may be useful for Software Engineers working on ML-related projects, or who want to build custom models, layers, and loss functions for their own projects.

See salaries and explore the career path for Software Engineer

Data Analyst

Data Analysts collect, clean, and analyze data to help businesses make informed decisions. This course may be helpful for aspiring Data Analysts who want to build custom models, layers, and loss functions in TensorFlow for data analysis and prediction. The course covers advanced TensorFlow techniques that are not commonly taught in introductory courses.

See salaries and explore the career path for Data Analyst

Product Manager

Product Managers are responsible for the planning, development, and launch of new products. This course may be helpful for aspiring Product Managers who want to learn how to build custom models, layers, and loss functions in TensorFlow to create data-driven products. The course covers advanced TensorFlow techniques that are not commonly taught in introductory courses.

See salaries and explore the career path for Product Manager

Quantitative Analyst

Quantitative Analysts use mathematical and statistical models to analyze financial data and make investment decisions. This course may be helpful for aspiring Quantitative Analysts who want to build custom models, layers, and loss functions in TensorFlow for financial modeling and prediction. The course covers advanced TensorFlow techniques that are not commonly taught in introductory courses.

See salaries and explore the career path for Quantitative Analyst

Technical Writer

Technical Writers create documentation and other materials to help users understand and use software and technology products. This course may be helpful for aspiring Technical Writers who want to learn how to build custom models, layers, and loss functions in TensorFlow to create technical documentation for ML products. The course covers advanced TensorFlow techniques that are not commonly taught in introductory courses.

See salaries and explore the career path for Technical Writer

Research Scientist

Research Scientists conduct research in various scientific fields and develop new theories and technologies. This course may be helpful for aspiring Research Scientists who want to build custom models, layers, and loss functions in TensorFlow for their research projects. The course covers advanced TensorFlow techniques that are not commonly taught in introductory courses.

See salaries and explore the career path for Research Scientist

Sales Engineer

Sales Engineers help customers understand and purchase software and technology products. This course may be helpful for aspiring Sales Engineers who want to learn how to build custom models, layers, and loss functions in TensorFlow to demonstrate ML products to customers. The course covers advanced TensorFlow techniques that are not commonly taught in introductory courses.

See salaries and explore the career path for Sales Engineer

Artificial Intelligence Engineer

Artificial Intelligence Engineers design, develop, and maintain AI systems. This course may be helpful for aspiring Artificial Intelligence Engineers who want to learn how to build custom models, layers, and loss functions in TensorFlow for AI applications such as natural language processing, computer vision, and robotics.

See salaries and explore the career path for Artificial Intelligence Engineer

Marketing Manager

Marketing Managers plan and execute marketing campaigns to promote products and services. This course may be helpful for aspiring Marketing Managers who want to learn how to build custom models, layers, and loss functions in TensorFlow to create data-driven marketing campaigns. The course covers advanced TensorFlow techniques that are not commonly taught in introductory courses.

See salaries and explore the career path for Marketing Manager

Machine Learning Architect

Machine Learning Architects design and implement ML systems and solutions. This course may be helpful for aspiring Machine Learning Architects who want to learn how to build custom models, layers, and loss functions in TensorFlow to create robust and scalable ML systems.

See salaries and explore the career path for Machine Learning Architect

Project Manager

Project Managers plan, execute, and close projects. This course may be helpful for aspiring Project Managers who want to build custom models, layers, and loss functions in TensorFlow to manage ML projects. The course covers advanced TensorFlow techniques that are not commonly taught in introductory courses.

See salaries and explore the career path for Project Manager

Business Analyst

Business Analysts use their knowledge of business processes and data to help businesses improve their operations. This course may be helpful for aspiring Business Analysts who want to build custom models, layers, and loss functions in TensorFlow to analyze business data and make recommendations for improvement. The course covers advanced TensorFlow techniques that are not commonly taught in introductory courses.

See salaries and explore the career path for Business Analyst