Blog posts

Deep learning has opened the door to tackling many real-world computer vision problems. But building and deploying Deep Learning models has always been a tedious task of labeling the entire dataset, finding the right hyperparameters, determining a data augmentation policy, and then deploying the model.

Masterful makes building and deploying Deep Learning models orders of magnitude easier. All you need to get started is to label a small fraction of your image data. Masterful’s meta-learner finds the optimal hyperparameters, augmentation policy, and trains the model both on labeled and unlabeled data. 

In this post, we’ll show how easy it is to solve a real-world computer vision problem using Masterful.

Semi-supervised learning (SSL) unlocks value in your unlabeled data, but it can be difficult to implement. Masterful can fully automate model training and apply SSL, but it also allows users to easily use SSL with their existing training code. We call this Simple SSL, and the benefits are less labeling and a more accurate model. In this post we'll show you how to try it.

Today we’re making a big leap forward in enabling new applications and insights for every enterprise that uses visual data. We’re releasing the next version of Masterful, the platform that automates model development for computer vision. Masterful now includes a low-code interface that enables developers to build models faster than ever before. Using a simple command-line interface (CLI), any developer can build production-grade models without needing to write code in TensorFlow or PyTorch. Our low-code interface also opens up CV development to an even wider universe of developers!

It’s no secret that just about any deep learning computer vision task can be improved with transfer learning. Transfer learning is a method where a model developed for one task is reused as the starting point for a model for another task. It leads to the question: how can we best reuse weights gathered from existing models trained on very large datasets to improve the performance and cut down development time on a new model for a smaller target dataset? In this post we’ll compare several approaches, provide experimental results, and show you how to easily incorporate transfer learning into your model development.

We’re excited to announce the v0.4.1 release of Masterful, the training platform for computer vision models!

Once your training runs become material in terms of wall-clock time and hardware budget, it's time to look at improving your batch size. If your batch size is too small, your training runs are taking longer than necessary and you are wasting money. And if your batch size is too large, you are training with more expensive hardware than you need.

Intro

Previously, we showed that throwing more training data at a deep learning model has rapidly diminishing returns. If doubling your labeling budget won’t move the needle, what next? Consider semi-supervised learning (SSL) to unlock the information in unlabeled data. 

Percent top-1 error reductions achieved when moving from Google Vertex to Masterful AI. 

We're thrilled to announce that the Masterful AutoML platform is now available on the Python Package Index for everyone to use. Installation is as easy as pip install masterful. You get the full power of our deep learning platform to train your models to peak performance without manual tuning. With Masterful, developers can focus on solving problems with ML instead of wasting time and brain cells running endless training experiments.