Category: Projects

JISC DigiFest Hackathon 2020

It’s been a year to the day since I published my last blog post reflecting on the DigiFest Hackathon of 2019. It’s been a busy year, but the opportunity to reunite the team for another Hackathon was one we couldn’t turn down!

As usual, the theme of this Hackathon is EdTech, organised by JISC. JISC are a UK non-profit, aimed at providing advice and resources in the tech sector, to schools and universities. Our team consisted of Nathaniel Read, Alex Lovett and Dan Tregoiing. We’ve completed a number of Hackathons together, and we’re all members (or alumni in my case) of the University of Hull, and Hull Computer Science Society (HullCSS). This event is for students and recent graduates, and JISC were happy to invite us to compete again.

A photo of the four hackathon team members in matching hoodies, on a stage
Team HullCSS (from left to right): Alex Lovett, Harry Gwinnell, Dan Tregoiing & Nathaniel Read

Nathaniel wrote a fantastic piece about the event as a whole, and I’d recommend you read it. This post will focus more on the technology behind what we built, and how it works.

Solution

oneGrade is our automatic marking solution, aimed at Computer Science teaching. Marking code is inherently complex problem given how many possible solutions exist. As such, we designed a solution to automate as much of the quantitative aspect of the marking process as possible.

The concept is simple; students submit their work (git or svn link) via a web interface or VLE integration, code is checked out and built into a container using Azure Kubernetes Service, and a suite of automated Unit, Interface, Integration etc testing can be performed. Azure Web Apps and Azure Container Service power this workflow, and results are delivered via the oneGrade web interface.

To setup an assignment, a lecturer must provide a Dockerfile that defines their assignment (e.g. an .NET Core image) and any number of testing scripts for their unit/integration/other testing framework that we support. This gives us the power of flexibility, allowing the platform to support any mix of languages and unlimited testing potential. This is all stored in the database against the assignment ID, and pulled out for use at container build time.

The Student interface is also simple, providing only a list of assignments. Assignments have either a submission box for a source URI, or a link to the report for their marks.

A flowchart of the solution architecture

The solution consists of two components, a React frontend providing our Instructor and Student UI, with an ASP .NET Core backend providing our Container Management services and testing environment. This hackathon was my first encounter with React, though I’m happy to say it was an experience I rather enjoyed. React proved a flexible and straightforward framework, and I was able to get going quickly. Most of my web experience is through the ASP.NET MVC Framework, with traditional JQuery.

The backend is responsible for taking a submission, building it into a container, and running the suite of defined tests. It does this using Azure Kubernetes Service (AKS). oneGrade will accept submissions, generate a worker process, and dispatch a request to AKS. Once AKS reports that the image is built and the container is running, the worker will run the defined testing, and store the results in the database. This provides a detached testing method, allowing for highly parallel operation.

Conclusion

Overall, the solution was fairly robust and solved the challenge we set for ourselves. There were a number of features we’d hoped to implement but simply ran out of time for, including a wider array of testing support, and using Azure Blockchain Service to store submissions for plagiarism detection. I’m pleased with how far we took the project during the Hackathon, and we may see some future iteration in our free time.

Our team was awarded the “Most Market Ready Solution” prize by JISC, for which we are immensely grateful. Competition was very strong with many incredible solutions being proposed, and both myself and the team would like to thank JISC for the opportunity to attend!

JISC DigiFest Hackathon 2019

The wonderful people over at JISC invited us along to their two day Hackathon, running alongside their DigiFest event. We pulled together a team of four, Nathaniel, Dan, Alex and myself and began coming up with ideas. They identified a number of key ideas they wanted us to build solutions for, including:

  • Student Wellbeing
  • Shaping the Curriculum
  • AI or Intelligent Agents to support learning
  • Intelligent Campus

Our idea was fairly straightforward. Our University (The University of Hull) has a whole host of services regularly accessed by students, often requiring student cards and logins to a variety of fragmented systems. Could we bundle all of these into a single source (A mobile app) and replace the 16,000+ physical plastic student cards with a Digital alternative. We put this under a combination of Intelligent Campus and Student Wellbeing, aiming to provide a platform that provides NFC powered mobile student cards, and a range of student services including help, security, timetables, travel etc.

We soon heard back that we’d been accepted, and a couple of months later we headed over to Birmingham for the event. JISC provides travel, hotel, food and everything we needed for the event, meeting on the first evening for dinner in a local restaurant. Much of the first night was spent throwing ideas around about how this might form, the kinds of services we wanted to offer while minimizing the amount of used space. The University currently uses an app called iHull, but this suffers from several issues around usability, and a lot of rarely used tiles on it’s home screen. We were looking for a fresh approach, something that aligned with the University’s brand identity more strongly.

As we made our way into the first official day of the event, we divided our four man team into tasks. Nathaniel worked on the UI Design (and some of the Xcode!), Alex and Dan worked on Backend services, and I was tasked with building the app in Xcode. Soon after, designs started flowing in from Nathaniel, and we were quickly constructing UI and linking together pages and Navigation Menus. Day 1 went by fairly successfully with a huge focus on getting the app looking presentable. We knew there wasn’t much time on Day 2, so we needed it to look good. We ended the day with a pretty satisfying looking app, and much of our backend services for login, weather and calendar information working, but not linked to our app.

At this point JISC invited us to their drinks reception, and we spent a couple of hours chatting and drinking with JISC employees and University delegates alike. Many interesting conversations were had, from AI Rabbit Robots to JISC Apprenticeships and more. Dinner followed in the most delicious pizza place, with our delegation filling almost the entire venue!

We finished off day 1 with a late night programming session, tying up our messy UI, fixing scaling bugs and deploying to TestFlight. We managed to tidy up most of the UI, leaving the evening with a reasonably function app.

Day 2 was split into two parts. Before lunch, and after. We spent much of the morning working on pulling the backend into the app we’d made, linking our login pages, UI, calendars etc to use the real data generated in the web platform. We then spent the afternoon building and tweaking our presentation. The presentation comprised mostly of screenshots of the app, using the strong brand identity present to produce and overall very professional looking presentation.

It was then time to present what we had made to a team of JISC Judges, the other teams, and a bunch of Delegates from the conference that were invited to attend. JISC’s own team of interns went first, followed by us, and the other teams. You can watch the presentation we gave here.

As the presentations wrapped up, the winners were announced, and we were delighted to be named the winners of this years event. All of the teams produced fantastic solutions, from interactive chatbots for campus services to a live French lesson using speech recognition. We’re incredibly proud to have been named the winner.

We’d like to thank JISC for inviting us to the event, and to everyone who was involved for making it a great competition.

We’re continuing to build out the app as a fun side project, and you can find more info on it here or contact me if you want to know more.

Getting started with Azure for students

I’ve had quite a few students tell me that they’d like to use a cloud platform, but they’re complex, hard to get started with, or expensive. I agree they can be quite confusing to a new user, but I’ve set out to make a simple ‘Getting Started’ guide for Microsoft Azure.

I’ve worked with the Azure platform for a couple of years now. As such, I’m starting to get an idea of what all the buttons do, and where all the settings are. However, Azure is a constantly evolving platform, and by the time you read this, it is very possible options have moved. Take these instructions as a guide, and maybe a search engine can help you find the rest.

We’ll start with how to get signed up for an Azure subscription as a Student. Azure has several layers of privilege behind its billing system. The main thing to know is every resource you create (VMs, App Services etc) is linked to a Subscription. Subscriptions have payment details linked to them, and determine how much you pay. If you use Azure in a professional capacity, you may use a Subscription setup by a CSP (Cloud Solution Provider) or some other entity. For our purposes, we’ll be setting up our own Imagine Subscription (Mine is called Dreamspark, yours may well be different too. As long as it’s not Pay-As-You-Go or Free Trial, you should be OK).

If you’re not a Student, or your School or University doesn’t have a partnership with Microsft (This is rare but it happens), you can sign up using the Free Trial here. Microsoft will give you $200 (at time of writing) to play  with. You can still use all of the free services, but if you’re not careful, you might end up paying for Standard Resources! Student subscriptions have a payment cap of £0 to stop them accidentally spending money!

Signing Up for Microsoft Imagine

First things first, you need to sign up for Microsoft Imagine, the Microsoft Student program. Imagine gives you access to a whole range of Microsoft services, and it’s worth having a browse around to see what’s available This will give you a link to use to sign up for Azure. If you wish to keep using your Imagine account, you’ll need to re-verify your Student status once a year. If you have done this already, you can skip ahead, otherwise read on to find out how to sign up.

  • Head over to Microsoft Imagine
  • Sign In in the top right corner (If you haven’t already) and click Imagine Account
  • Microsoft Account Menu
  • Next, you’ll need to verify your Student status using the pane  on the left. If you’ve already done this, no need to do the next step.
  • Student Status panel
  • Select how you’d like to verify your Student status. This will vary based on your institution (I used my Student email address)
  • Finally, complete your verification (In my case, using the email from Microsoft)

Get your link to the Azure Sign up

Now we’ll get signed up for Azure using our Student status. As of writing, this link will get you $100 of credit, and access to a bunch of Always Free and Free for 12 Month services.

  • Head to this Microsoft Imagine page
  • Click the ‘Register Now’ button. If you don’t see it, you may not be signed in, or you may not have verified your Student status
  • Register Now button
  • You will then be redirected to Azure to complete the process.
  • Complete the Identity Verification presented
  • Then. complete the Card Verification. If you use any chargeable resources, this is the card they will be charged to!
  • Finally, complete the Agreement
  • You should then be left in the Azure dashboard
  • Azure Portal

Now what?

Feel free to explore the Azure dashboard and the resources available. Be aware that if you create any, they may cost money, and be charged to your card.

Over the coming weeks, I’ll be writing a few guides for projects you can do to get started on Azure. These will only use Free resources, so you can try them out too!

 

How’d it go? Has anything changed? Ran into issues? Got an idea for a project you want to build in Azure?

Leave me a comment, or Contact Me!

ImgBot – A Discord Machine Vision project using the Azure Vision API

I’ve started a project this week called ImgBot. He’s a little Discord bot that uses the Vision API from Azure to recognise images posted to our HullCSS Discord server. As of now, he’s not super clever, only using the basic image APIs to gather data about a given image, but in the coming months I’m hoping to expand into a full Custom Vision model, for more advanced recognition.

For some idea of how it works, I’ve included a screenshot of the current version. In future, I’m hoping to allow users to give the bot feedback, to improve the model.

Discord Screenshot showing ImgBot recognising Nicholas Cage

ImgBot recognising Nicholas Cage

It’s also possible to get more advanced information from the bot, having it return the full JSON object representing the image.

Discord Screenshot showing ImgBot giving a full JSON object for the Nicholas Cage image

ImgBot giving more detail on Nicholas Cage

 

I might look to Open Source this project later down the line, once the code is cleaned and organised. It’s currently fully functional however a little messy.

Let me know what you think and how you think I could improve the bot. Interested in testing it out? Drop me a message!

Google HashCode @ Hull 2018

Yesterday (1st March 2018), a number of HullCSS students got together to compete in Google’s HashCode competition. Everyone managed to submit a solution and gain points, and great fun was had by all.

The Event

HullCSS (The Hull Computer Science Society) ran a HashCode Hub in the Fenner SuperLab, where teams from all over the University came together to compete. It was the first time HashCode was run at the University of Hull, and all students were welcome.

All teams managed to submit a working solution, with the lowest point total being 10 Million, and the highest being 45 Million. All were impressive solutions, with each taking a different approach to solving the problem. Everyone relished the challenge, and it was interesting to see all skill levels working together.

All teams were eagerly watching the scoreboard throughout the event. Everyone wanted to make it above the top 1000 teams (top 25%).

My team consisted of Josh Taylor, Alexander Rossa, and myself, and was named Gravity Gun. We kicked off at 5.45pm with the release of the problem (which you can find here), and got to work planning out our algorithm.

The Solution

We decided to take a Supervisor/Worker approach, with our main program acting as a coordinator of rides, and the cars doing much of the route finding/calculation themselves.

A basic set of properties were given to the car, to allow it to know about itself, such as its position, the rides it has completed, and the current global step it is on. This allows the supervisor to ask the car how long it will take to complete a given ride. The car is capable of calculating at what points it will pick up the customer, and arrive at the destination.

The supervisor knows more generalised information about the problem, such as the size of the grid, the list of cars and rides to complete, and the number of steps in which the rides have to be completed. It prioritised the list of rides, then asks each car how long it will take to complete the ride. In this version the prioritisation is very basic, and is only ordered by the end step in ascending order. It then gathers the results, and immediately removes any result that will complete after the maximum finish step of the ride. If the ride won’t complete the ride in time, we don’t even want to attempt it, as we could be using this time to get more points.

Next, it will check to see if any of the cars can achieve the bonus points. If they can, it will pick one of these cars, otherwise it will pick any car. It will then pick the car that reported the least number of steps to complete the job. This provides a fairly optimal path for completing rides (though clearly not the best!).

Conclusions

This model is fairly good, achieving 45 million points, though not without its faults. The top team managed to achieve over 49 million points. Given more time (there was a fair amount of time finding and fixing silly logic errors), we’d look to improve the prioritisation algorithm, looking at balancing the choice of going on a long journey, vs doing multiple short ones.

The competition was great fun, and I’ll certainly look to do it again next year. I’d encourage any student who wants experience with a real world, complex programming problem to compete! It offers something drastically different from  a University module, and it can be really rewarding to build a working solution. As a student, the challenge was really engaging, but not too difficult as to be unsolvable.

I’ll definitely be encouraging HullCSS to run the event again next year. If you’re in Hull and want to get involved, let me know. If there is enough interest, it may be that the Hub is opened up to the whole city!

Links

Want to know more about HashCode or get involved? You can find some info here:

https://hashcode.withgoogle.com/

 

Want to see my solution? Check it out on GitHub! (We were really tired, and were rushing when we wrote this, apologies in advance!)

https://github.com/HarryGwinnell/HashCode2018

 

Want to have a go yourself? Find the PDF problem here! And the inputs are here.

 

Why not check out my other projects?