Crop Finder

Inspiration

Farmers spend hours upon hours doing independent research to find the ideal crops for their soil. We want to help them do it quickly and accurately.

What it does

It takes 7 inputs including the nitrogen level, the potassium level, etc. of a soil and recommends the 5 best crops for this soil. It gives a “fitness score” for each crop.

How we built it

The front end was done with HTML, Javascript, and CSS. The backend was done with Python and Pandas library. The backend framework we used was Django.

Challenges we ran into

We tried to use a dictionary at first with key value pairs for the crop and its mean data; however, we struggled to iterate through the dictionary and find the smallest differences between them. From here, we just chose to make everything a list within a list. We also had trouble implementing the first algorithm, which was complex and inaccurate, and after much hassle we substituted that with a simpler and more accurate one.

Accomplishments that we're proud of

We are proud of how we learned to read csv files with Pandas and Python. We’re also proud for having learned to effectively collaborate on such a complicated project.

What we learned

Data analytics is very difficult and requires a good understanding of recursion and data structures.

What's next for Crop Finder

We want to factor in the user’s GPS location to give more accurate evaluation of the fitness of each crops. We want to add more types of crop to our database so the users can have more choices. We also want to use our existing algorithm to show how a farmer can adjust their soil to fit a crop’s growth condition.