Tensorflow with Images
Suitable for students in Middle School and High School (Grades 6-12)
Learn image classification with TensorFlow - the state-of-the-art Image Processing techniques available today.
In this advanced class, students with some AI experience in images learn how to build image processing pipelines in Python using Tensorflow - the State of the Art image classification techniques available today. They will build their own custom project in images using state-of-the-art deep learning models.
Learn Artificial Intelligence - a new technology that is shaping our world!
Students enrolling are expected to have an introductory deep learning class in images (M3 or Summer Camp) and basic python background, which could be an entry-level python course with us (PA1 or Summer Camp).
Why learn TensorFlow and Natural Language Processing
Python is a language in very high demand because of its versatility and penetration into just about every industry. TensorFlow, built by Google, is a state-of-the-art deep learning platform. Image classification within deep learning is the Artificial Intelligence field that deals with images, be it natural, synthetic, or medical. Automated processing of images is used widely everywhere in the real world for reducing human labor and subjectivity.
Why join our AIClub Summer Camp?
-
Designed and taught by PhDs and AI Experts
-
Proven - hundreds of K-12 students have built custom projects. Some have won STEM and Innovation competitions
-
Only workshops where students can build Python-powered AIs and publish them online! Students love building AIs and learning how they work
-
Fun! Students build many programs and complete a project of their choice.
Please see our brochure for more info about our programs!
Description
This course will reinforce core programming concepts (input/output, conditionals, loops, data structures, etc.). Students will then learn the basics of TensorFlow in Python, including how to prepare image data. They will learn about various types and formats of images and how to process them to be able to build TensorFlow models. They will learn about several state-of-the-art models including ResNet, MobileNetV2, etc. They then combine their newfound skills to create a custom project - an image processing project of their choice. They will present and demo their project to the class at the end of the course. Parents will also receive a video of their child's presentation and students can also write a blog about their project that will be showcased on the AIClub website.
What Students Take Away
-
Review of core programming concepts - flowcharts, input/output, loops, conditionals, data structures. Introduction to data preparation mechanisms for image data
-
Transfer Learning - how to transfer the learning of a state-of-the-art model for your own custom data. How to process large amounts of data by splitting the TensorFlow pipelines.
-
How to combine Python, Image Data Preparation and Deep Learning to create an intelligent app that can communicate with users and classify images.
-
Build a custom Image Processing project and application in Python.
Schedule
Duration: 5 sessions / 3 hours per session
We offer a range of dates and times to accommodate busy schedules.
Also - note that since we use entirely online tools, if a student must miss a class, it is easy for them to do the required work at home. We provide materials for missed classes. We do ask however that the student attend the first and last class since this is needed for them to get oriented and also complete their custom project.