renangdc / final_project

GOAL:

On the final Project from Ironhack, we have to build 3 Prediction models and create a presentation.

Description:

Welcome, this is an python app created using streamlit. The app is a Phone Recommender, based on 4 of the most important features for most of the users. Such as battery capacity, screen sizes in inches, camera quality and how well it performs.

Notebooks explanation:

Analysing_and_transform_for_recomb: Used for a quick analysis of the data set and to performe such transformations for exemple the currency of the data set on the list of prices to euros.

Clustering_data: Used to cluster the data using kmeans in order to understand the needed amount of clusters for each chosen category.

Re_clustering: Used to apply the best amount of clusters for each ranges of the chosen categories, keep in mind that for each cluster there are differente amounts of columns.

Recommenders: Used to check each of 2 recommender functions in order to see the outputs and choose the desired one.

Recommender_final: The final choosen recommender, used to build the app.

Applying_rf: Since it was part of the final project on this notebook the Random Forest model was used to predict the prices of the phones, also the Grid Search was used to refine the best parameters for the Random Forest.

Data set source: https://www.kaggle.com/datasets/pratikgarai/mobile-phone-specifications-and-prices