The last 2 weeks in the #community-events:fastbook reading group have taught me A LOT about collaborative filtering. I tried to build my own movie recommender and documented what I learned in the process in these posts:
- Building a Movie Recommender - Part 1
- Building a Movie Recommender - Part 2
I’m planning to now convert this model into a usable web application.
That would mean creating an API that returns recommendations using the model (using something like Flask or FastAPI), deploying it on the cloud somewhere (maybe Heroku or as an AWS Lambda) and building a simple web application that allows users to rate movies and see recommendations (maybe using React/Angular or something similar)
Is anyone interested in working on this together? 
10 Likes
Hi @ravimashru , I would love to contribute to this. Let’s discuss on how to start working on this together.
2 Likes
Awesome! Looking forward to building something cool together 
Yaayy! I’m really looking forward to learning more from you in this mini-project. 
1 Like
Amazing the mentor @amanarora himself wants to join us 
2 Likes
Keep us posted on how it goes We’d love to learn from your experience 
2 Likes
Hi Everyone!
I am interested in working on a mini-project on Recommendation Systems. We can try out the tutorial and expand on that work. If anyone is interested, lets discuss regarding the same!
2 Likes
Hi @ravimashru, I am also interested in joining this. I had personally tried building a webapp for deploying a for our image classifier models with VueJS + Flask. Yet was not able to complete it successfully yet as I faced, connecting both frontend and backend.
Also if this is something, which can be made generic to be sort of a small frameworks in future. So if someone want’s to build a recommendation engine for someother usecases, they can use the frontend & backend and just plugin their ML model?
2 Likes
Hey @ravimashru lets connect over a call.
2 Likes
Cool! I love the idea of trying to make this something reusable.
1 Like
Roger that! I can set up a call.
Any date and time preferences? One option would be Wednesday 8:00 am IST (Tuesday 7:30 pm PT).
1 Like
Ya that should work. Let’s connect. Thanks.
Just scheduled a call. See you all then!
Wednesday, 1 September · 8:00 – 9:00am IST
Video call link: https://meet.google.com/xtv-cgir-kbs
1 Like
@kurianbenoy-aot and I had the first call today to kick this off 
Here are some things we discussed:
Information
Decisions
- The tech stack we’ll use is: FastAPI + VueJS
- Working model: mostly working asynchronously with 1 - 2 check-in calls a week to keep us on track
Action items
- Train a model on the MovieLens 25M Dataset.
- Learning & upskilling (FastAPI & VueJS)
- Try out Deta as a possible free hosting option for the backend
Next meeting
Tentative date/time: Saturday, 4 September. 7pm IST (if it works for everyone)
Tentative agenda: create a skeleton for the different parts of the project, create repo on GitHub.
4 Likes
This is incredible!
May I suggest looking at NVIDIA Merlin ecosystem for some ideas? The team has been winning RecSys comps consistently and usually open source ideas on here. You might find something interesting to experiment.
The blog on medium also tends to have great suggestions
3 Likes
Thanks! We’ll check these out.
Hey guys. This is a great project. I am currently watching this space actively for ideas
I have also been working on a similar recommendation system but for books
But was stuck at the putting it in production part. I’m currently hosting at Streamlit but for some reason my app keeps on crashing because I’m using too much resources after some time.
I have learned about sparse embeddings @bhutanisanyam1 and Deta for hosting @ravimashru so you guys have been very resourceful
Planning on joining in the other call
3 Likes
