We do not wish to diminish the motivation of any pupil under the age of 16. Any aspiring medic under the age of 16 can feel reassured that the resource will be available for them when they turn 16. Our intention is to continue adding to the platform and to provide this resource to aspiring medics long-term. The content of the experience has also been carefully designed for pupils aged 16 and over. It is not possible for us to do this and so, at this time, pupils aged under 16 unfortunately aren’t able to register to access the platform. Legally the RCGP would be required to gain parental/guardian consent for any pupils under 16 who share their personal data with us. A link is then sent via email to access the platform and they are prompted to create a login (using their name and email address) directly on the platform, which is housed by a company called Koantic (see the section below called 'What data do you collect and why?' for more information. Habitat supports PyTorch, is easy to use, and is open source.Sadly not, to gain access to the platform, pupils are required to complete a registration form, providing some personal information. We implement our technique into a Python library called Habitat and find that it makes accurate iteration execution time predictions (with an average error of 11.8%) on ResNet-50, Inception v3, the Transformer, GNMT, and DCGAN across six different GPU architectures. We make predictions by scaling the execution time of each operation in a training iteration from one GPU to another using either (i) wave scaling, a technique based on a GPU's execution model, or (ii) pre-trained multilayer perceptrons. Our technique exploits the observation that, because DNN training consists of repetitive compute steps, predicting the execution time of a single iteration is usually enough to characterize the performance of an entire training process. In this work, we present a new practical technique to help users make informed and cost-efficient GPU selections: make performance predictions with the help of a GPU that the user already has. However, choosing which GPU to use is challenging both because (i) there are many options, and (ii) users grapple with competing concerns: maximizing compute performance while minimizing costs. Deep learning researchers and practitioners usually leverage GPUs to help train their deep neural networks (DNNs) faster.

Runtime of doctors in training videos