Custom networks in MXNet on Julia – part 2

Hi, Albin here again! Part of the work I do at LakeTide is making our machine learning models more computationally efficient. For instance, sometimes the neural networks we develop end up being big enough that they don’t fit on GPUs anymore. Other times we just...

Setting up GPU passthrough with KVM on Fedora

In a recent LakeTide project, we wished to emulate a 12-server rack by setting up VMs on our beefy local workstations. The aim was to include one VM with 1 GTX 1080 and another with 4 GTX 1080s. Ultimately, the combination of Fedora, KVM, and nvidia-docker made...

DC/OS and the Mesos scheduler

At LakeTide, we constantly evaluate technologies that can deploy computation-heavy code to resource clusters and DC/OS is one of the heavyweight contenders in the container orchestration space right now. Though it has a smaller community than Docker, Kubernetes,...

Custom networks in MXNet on Julia – part 1

  This is a small example of how to build a MNIST classifier with MXNet and Julia using executors with customized optimizer and accuracy functions. MXNet is a deep learning framework that is very well suited for parallel computations over several GPUs, i.e. it’s a...

LakeTide @ Nordic Data Science Summit

A few weeks ago, the inaugural Nordic Data Science Summit (NDSS) took place in Stockholm. With more than 300 attendees, this “first” conference was jam-packed with passionate smart people, interesting sessions, and engaged partners, providing a very...