Announcement Paperspace Closes $13M to Fuel Growth We are excited to announce that we have closed a $13M Series A from SineWave Ventures, Battery Ventures, Intel Capital, with follow-on from Initialized Capital. This latest round brings our total funding to $19M.
Announcement Teams! Many of our users use Paperspace as part of a team (at a company, university, etc.) working collaboratively together on projects. A highly requested feature was to be able structure those teams inside
Volta Mixed Precision Training with NVIDIA Volta A quick overview of what is and what are the capabilities of mixed precision training with the NVIDIA V100 GPU card.Volta is one of the latest GPU architectures developed by NVIDIA. Volta
Tutorial Generating an interactive Pix2Pix model with Gradient° and ml5.js This post will go through the process of training a generative image model using Gradient° and then porting the model to ml5.js, so you can interact with it in the browser. For
CI/CD CI/CD for Machine Learning & AI The ecosystem for developing modern web applications is incredibly rich. There are countless tools for delivering a modern web app to production, monitoring it's performance, and deploying in real-time. These tools are so
Gradient Introducing GradientCI our new friendly CI/CD bot for Machine Learning and AI pipelines We believe that Machine Learning is in a great spot for ... Introducing..GradientCI our new GitHub integration that makes running your ML jobs easier than ever Install on your private github repos here:
Series: Data Augmentation Data Augmentation for Bounding Boxes: Rethinking image transforms for object detection How to adapt major image augmentation techniques for object detection purposes. We also cover the implementation of horizontal flip augmentation.
Series: Data Augmentation Data Augmentation for Bounding Boxes: Scaling and Translation We implement Scale and Translate augmentation techniques, and what to do if a portion of your bounding box is outside the image after the augmentation.
Computer Vision Data Augmentation for Bounding Boxes: Rotation and Shearing This is part 3 of the series where we are looking at ways to adapt image augmentation techniques to object detection tasks. In this part, we will cover how to implement how to rotate and shear images as well as bounding boxes using OpenCV's affine transformation features.
Series: Data Augmentation Data Augmentation For Bounding Boxes: Building Input Pipelines for your detector Previously, we have covered a variety of image augmentation techniques such as Flipping, rotation, shearing, scaling and translating. This part is about how to bring it all together and bake it into the input pipeline for your deep network.
Announcement The New Paperspace Community We're always amazed by what our users are building with Paperspace and felt it was important to provide a space where people can share ideas, ask questions, learn new tools and techniques and
Paperspace Cloud reliability & performance improvements 💪 We've come a long way as a GPU cloud, supporting over 100k users and continuing to scale rapidly. At times, our growth has imposed a burden on our systems, sometimes in ways we
Gradient Gradient update (7/18/2018) Gradient° has been updated in response to a ton of feedback from the community. Here's a roundup of some of the things we've added recently: Product release notes can be found here and
Quilt Reproducible machine learning with PyTorch and Quilt In this article, we'll use Quilt to transfer versioned training data to a remote machine. We'll start with the Berkeley Segmentation Dataset, package the dataset, then train a PyTorch model for super-resolution imaging.
Series: Optimization Intro to optimization in deep learning: Busting the myth about batch normalization Batch Normalisation does NOT reduce internal covariate shift. This posts looks into why internal covariate shift is a problem and how batch normalisation is used to address it.
Machine Learning Creating your own style transfer mirror with Gradient° and ml5.js In this post, we will learn how to train a style transfer network with Paperspace's Gradient° and use the model in ml5.js to create an interactive style transfer mirror. This post is
Series: Optimization Intro to optimization in deep learning: Vanishing gradients and choosing the right activation function An look into how various activation functions like ReLU, PReLU, RReLU and ELU are used to address the vanishing gradient problem, and how to chose one amongst them for your network.
Training an LSTM network and sampling the resulting model in ml5.js In this post, we will learn how to train a language model using a LSTM neural network with your own custom dataset and use the resulting model inside ml5.js, so you will
Series: Optimization Intro to optimization in deep learning: Momentum, RMSProp and Adam In this post, we take a look at a problem that plagues training of neural networks, pathological curvature.
Series: Optimization Intro to optimization in deep learning: Gradient Descent An in-depth explanation of Gradient Descent, and how to avoid the problems of local minima and saddle points.
Gradient What's new in Gradient? We've been hard at work developing Gradient° into a robust and scalable deep-learning platform. Here's a roundup of some of the things we've added recently: Product release notes can be found here and
Tutorial Vectorization and Broadcasting with Pytorch The performance gains derived from running your machine learning code on a GPU can be huge. But GPUs are optimized for code that needs to perform the same operation, thousands of times, in
GAN Building a simple Generative Adversarial Network (GAN) using TensorFlow Generative Adversarial Networks or GANs are one of the most active areas in deep learning research and development due to their incredible ability to generate synthetic results. In this blog, we will build out the basic intuition of GANs through a concrete example.
Series: YOLO object detector in PyTorch How to implement a YOLO (v3) object detector from scratch in PyTorch: Part 1 Tutorial on building YOLO v3 detector from scratch detailing how to create the network architecture from a configuration file, load the weights and designing input/output pipelines.
