To wide fanfare, Google in February published its internal Machine Learning Crash Course program originally designed as a resource for its employees to learn about AI. Learn with Google AI is now adding a new practicum focussed on image classification.
These resources are aimed at helping developers learn more and integrate machine learning into their own services and applications. Feedback Google has received since February revealed that learners want more practical insights about deploying AI into products.
In particular, you want to learn from teams who have built and deployed ML models. What challenges and successes do product teams encounter? How do they problem solve, and what solutions work best?
As a result, the Google AI team worked with the company’s image model experts to develop the “Machine Learning Practicum on Image Classification.” The interactive course walks developers through the basics of how image classification functions and later instructs them on how to build a convolutional neural network. It also includes insights from Google Photos.
This hands-on practicum contains video, documentation, and interactive programming exercises, illustrating how Google developed the state-of-the-art image classification model powering search in Google Photos. To date, more than 10,000 Googlers have used this practicum to train their own image classifiers to identify cats and dogs in photos.
Google hopes to release more Machine Learning Practica in the future.