The NVIDIA Transfer Learning Toolkit is now NVIDIA TAO Toolkit. It’s important for the model to make accurate predictions when using a deep learning model for production. How efficiently these predictions happen also matters. Examples of efficiency measurements include electrical engineers measuring energy consumption to pick the best voltage regulator, mechanical engineers timing inference…
]]>In part 1 of this series I introduced Generative Adversarial Networks (GANs) and showed how to generate images of handwritten digits using a GAN. In this post I will do something much more exciting: use Generative Adversarial Networks to generate images of celebrity faces. I am going to use CelebA [1], a dataset of 200,000 aligned and cropped 178 x 218-pixel RGB images of celebrities.
]]>You heard it from the Deep Learning guru: Generative Adversarial Networks [2] are a very hot topic in Machine Learning. In this post I will explore various ways of using a GAN to create previously unseen images. I provide source code in Tensorflow and a modified version of DIGITS that you are free to use if you wish to try it out yourself. Figure 1 gives a preview of what you will learn to do in…
]]>Today we’re excited to announce NVIDIA DIGITS 5. DIGITS 5 comes with a number of new features, two of which are of particular interest for this post: In this post I will explore the subject of image segmentation. I’ll use DIGITS 5 to teach a neural network to recognize and locate cars, pedestrians, road signs and a variety of other urban objects in synthetic images from the SYNTHIA dataset.
]]>