Olga Petrova - Scaleway Blog - European cloud provider

Smart data annotation for your computer vision projects: CVAT on Scaleway

In this article we are going to look at how to set up a data annotation platform for image and video files stored in Scaleway object storage, using the open source CVAT tool.

How to make a demo of your machine learning model with gradio

As an AI engineer or a data scientist, you are probably familiar with the following scenario: you have just spent weeks developing your new machine learning model, you are finally happy enough with

Supervised Machine Learning done right: looking beyond the labels

Sometimes machine learning projects require additional labels to perform in the wild. Learn what these are, and how Smart Labeling can help you get them! Before we get into this, let me...

Supervised Machine Learning done right: getting the labels that you want

Best manual data annotation practices for your machine learning projects. “Garbage in, garbage out” is an expression that Machine Learning practitioners are very familiar...

cloud

Doing AI without breaking the bank: yours, or the planet’s

A 2018 study by OpenAI showed that the amount of compute power needed to train state-of-the-art AI models was doubling every 3.4 months. Such exponential growth translated into an astounding 300,000x-fold

AI

VAE: giving your Autoencoder the power of imagination

In this article we will take a detailed look at the Variational Autoencoder: a generative model that is based on its more commonplace sibling, the Autoencoder. Stay tuned for the PyTorch implementation in the next post!

GPU

Moteur de recommandation : dois-je utiliser un CPU ou un GPU ?

Dans un monde dominé par "la data", les GPU sont le hardware de référence pour le deep learning. Mais qu’en est-il des tâches "simples" ? Dans le cas un moteur de recommandation un GPU est-il pertinente?

GPU

CPU or GPU for your recommendation engine?

In today's data-driven world, GPUs are the hardware of choice for training Deep Learning models. What about tasks that do not involve artificial neural networks? For instance, is there a benefit to using a GPU for making product recommendations? Continue reading to find out!

AI

Active Learning, part 2: the Practice

This blog post is the continuation of "Active Learning, part 1: the Theory", with a focus on how to apply the said theory to an image classification task with PyTorch.

AI

Active Learning, part 1: the Theory

Active learning is still a niche approach in machine learning, but that is bound to change. After all, active learning provides solutions to not one, but two challenging problems that have been poisoning the data scientists' lives. Those are, of course, the data: its (a) quantity and (b) quality.

AI

Semi-Supervised Learning with GANs: a Tale of Cats and Dogs

Semi-supervised machine learning is a solution when labeled data is scarce. In the article we introduce a semi-supervised Generative Adversarial Network for image classification. Read on to find out how to get a 20% increase in accuracy when distinguishing cats and dogs with only 100 labeled images!

Fresh from the arXiv: Nov 11 to 15

Discover how networks can learn from one another using unlabelled data, why you should think before you prune, and whether you can make a network forget what it saw.

Fresh from the arXiv: Oct 21 to 25

Learn how Generative Adversarial Networks (GANs) can help you when you do not have enough labelled data for your supervised machine learning problem, and find out what is good about a bad GAN.

Fresh from the arXiv: Oct 14 to 18

A selection of machine learning preprints from the week of 14.10.19 that the Scaleway AI team found interesting.

Fresh from the arXiv: Sep 30 to Oct 4

A selection of machine learning preprints from the week of 30.09.19 that the Scaleway AI team found interesting.

Fresh from the arXiv: Sep 23 to 27

A selection of machine learning preprints from the week of 23.09.19 that the Scaleway AI team found interesting.

Fresh from the arXiv: Sep 16 to 20

A selection of machine learning preprints from the week of 16.09.19 that the Scaleway AI team found interesting.

Fresh from the arXiv: Sep 9 to 13

A selection of machine learning preprints from the week of 09.09.19 that the Scaleway AI team found interesting.

Understanding text with BERT

How to build a machine reading comprehension system using the latest advances in deep learning for NLP.

Fresh from the arXiv: Aug 26 to 30

A selection of machine learning preprints from the week of 26.08.19 that the Scaleway AI team found interesting.

AI

Natural Language Processing: the age of Transformers

How to build a machine reading comprehension system using the latest advances in deep learning for NLP.

Scaleway

Introducing GPU Instances: Using Deep Learning to Obtain Frontal Rendering of Facial Images

It’s article time! Check how we used our GPU instances to obtain frontal rendering of facial images with Deep Learning.