n this article, we looked in depth at ControlNet, a new technique for imparting high levels of control over the shape of synthesized images, and demonstrate how to run it in a Gradient Notebook
In this article, we explored a broad overview of epistemic uncertainty in deep learning classifiers, and develop intuition about how an ensemble of models can be used to detect its presence for a particular image instance.
In this article, we looked at the novel VALL-E TTS model, and showed how to train it within a Gradient Notebook using Libri Light and our own voice recordings.
In this article, we took a brief look at uncertainties in deep learning. Thereafter, we took a more keen look at aleatoric uncertainty and how convolutional autoencoder can help to screen out-of-sample images for classification tasks.
In this article, we examine typical computer vision analysis techniques in comparison with the modern CLIP (Contrastive Language-Image Pre-Training) model.
We compare the performance of using different techniques for handling Missing At Random datasets in building predictive models. We also examine how these techniques affect the predictive performance of machine learning models.
In this blog post, we examine what's new in Ultralytics awesome new model, YOLOv8, take a peak under the hood at the changes to the architecture compared to YOLOv5, and then demo the new model's Python API functionality by testing it to detect on our Basketball dataset.
When it comes to image synthesis algorithms, we need a method to quantify the differences between generated images and real images in a way that corresponds with human judgment. In this article, we highlight some of these metrics that are commonly used in the field today.
In this article, we will understand how to use various tools to maximize GPU utilization by finding the right batch size for model training.
