Food Identifier

Inspiration

We wanted to learn more about ML and while brainstorming we were trying to see how we could implement ML into everyday in a way that would make life a little bit easier.

What it does

At the moment we have a created a data model as well as a streamlit web application that has a file upload system. Due to time constraints we were unable to make more progress.

How we built it

We found a dataset form kaggle that had pictures sorted of different fruits and vegetables and then we created a data model on the notebook of kaggle. Then downloaded the model and created a python file to convert the images to be able to use the model to identify the food.

Challenges we ran into

We spent a lot of time learning new concepts as well as downloading/installing new software. We also ran into a lot of error that took time to fix and the most recent one was a connection error because of an issue with the ports for flask.

Accomplishments that we're proud of

We were really that we were able to create a data model because it was new to the both of us and we were able to learn more about the backend process of how ML works and the way it is useful.

What we learned

We learned how to use datasets from the internet to create data models as well as how to deploy a web app using streamlit.

What's next for Food Identifier

Hopefully, the next steps is to be able to get app to work in a way it can identify multiple objects in the same picture and list them. Then is will give recipes that have the ingredients listed by the data model. It will hopefully also have text to speech for ease of use.

Built With

• kaggle
• python
• streamlit
• tensorflow