force-tribble-chef

An webapp to predict your competitive programming ratings based on your platform.

Force-Tribble-Chef

This project uses a dataset of CodeChef and Codeforces usernames with their ratings in respective platforms. I cleaned the data and ran some EDA, then used Linear regression to find the relationship between the two platforms. I used this to predict the rating of a user on one platform given their rating on the other platform.

To get started

You'll need Python 3, NumPy, Pandas, and Scikit-learn installed. Once you have those, you can run the project by following these steps:

• Clone the repo:

git clone https://github.com/Pin4sf/Force-Tribble-Chef.git

• Change directory to the project directory:

cd force-tribble-chef

• Open the python file and see the EDA

Results

The analysis showed that there is a strong correlation between the ratings of CodeChef and Codeforces users. This means that you can use a user's rating on one platform to predict their rating on the other platform.

The prediction model was able to predict the ratings of users with an accuracy of 80%. This means that the model was able to correctly predict the rating of a user 80% of the time.

I have created an website based on the results of machine learning model from python file you can check out the website from here[https://pin4sf.github.io/force-tribble-chef/]

Screen Shots

Future work

This project could be extended in a number of ways. For example, the dataset could be expanded to include more users. The model could also be improved by using a different machine learning algorithm.