Much of the reason that deep learning has become popular and successful in recent years is the availability of vector processing with flexible GPU compute. As a Machine Learning Research Group we utilize GPU acceleration to conduct all of our experiments. As a part of the Tutorials for the Canada-France-Iceland Workshop on Machine Learning and Computer Vision we needed a GPU accelerated environment that could be used by 50 or more people.
This post provides very simple instructions for connecting to the uog-wifi-secure network at University of Guelph (specifically proven working in 2018). These instructions can likely be adapted to any WPA-enterprise network utilizing EAP-PEAP (MSCHAPv2) and any device using wpa_supplicant. From raspbian you can edit two files,
/etc/network/interfaces (to enable our interfaces, i.e. turn them on) and
/etc/wpa_supplicant/wpa_supplicant.conf (for entering the required parameters).
- MB: GIGABYTE B360M D3H
- CPU: Intel Core i7-8700K 3.7 GHz
- Cooler: Cooler Master (all in one watercooler) RL-S12M-24PK-R1
- SSD1: ADATA Premier SP550 120GB
- SSD2: Samsung SSD 840 Series 120GB
- Storage: 5TB of HDD
- Memory: CORSAIR Vengeance LPX 32GB (2 x 16GB)
- GPU1: ASUS GTX970 STRIX
- GPU2: Sapphire Radeon HD 7950
- PSU: Thermaltake 750W SP-750PCBUS 750W
- CASE: Thermaltake Core V21 Black Extreme Micro ATX Cube Chassis
- OS: Arch Linux/Windows 10 VM
Pokémon Go was released on July 6th, and in Canada has July 17th. Before it was released in Canada I was attempting to find ways to trick the Play Store in order to get it early. If you search for these type of region bypasses on popular search engines you get links to sites with sketchy downloads of Android apks. I eventually gave up trying to get it from the official Play Store while a huge number of Canadians already found ways of getting it.
To the point. I will not be playing Pokémon Go now that it is available in Canada. I provide the following reasons.
Graduate from first year, stop doing the work for the computer
Computers are really good at doing repetitive tasks so lets address some of the time saving and more professional ways of writing code
Some of the people in our department will disagree with some of my decisions here but I'm choosing these based on platform independence, wide spread use, and potential. The idea of this section is to get you into self learning in computer science. Self learning is probably one of the fundamental concepts in industry. With the competition among top performers in industry if you do not learn to self learn but still get good grades you will probably do well, but if you really want to be the best of the best you need to self learn.
As an introduction I would like to assert that nothing said in this post covers all situations and view points that are present. The topics presented are what advice I can draw from personal experience as a Helpdesk TA and with discussions with other upper year students. The list is ordered in my order of importance.