Gemma-GB

• 

  GitHub

• 

  Google AI Studio

Inspiration

I wanted a fun project and I really liked the concept of reccomending movies based on the a really good prediction. Particularly, I liked the idea of using Ghibli movies more rather than other broad genres.

What it does

It takes user input as either genre or the previous ghibli movies they watched to reccommend something similar or maybe a suprise reccomendation? It can take only either of two inputs for now.

How we built it

I built it in google generative studio using google gemini API. I used prompts and conversed with the model too. I added character based tones for better and engaging interction with the user end.

Challenges we ran into

I got lots of bugs, I am still working on it. The code needs to be optimized too for a better performance.

Accomplishments that we're proud of

I got to use prompts with the model I trained on. As a first project, maybe it is not too much of a working or functional stuff but I tried at least.

What we learned

I learned about lots of different libraries and modules. I learned about Streamlit platform and it was easy to deploy but the code had bugs, so it was not possible. I also learned about tunneling for this app deployment using python that I am yet to do.

What's next for Gemma-GB

It needs to work first of all. And I want to add up some interactive elements using background sounds and effects. The project might be deployed via Streamlit.

Built With

• google
• google-gen-ai
• python