Projects
Coffee Arabica Price Prediction
Application of a Long Short-Term Memory (LSTM) machine learning model for daily price prediction (in BRL) of Arabica coffee in Brazil, using ten years of historical data.
​
The model successfully learned the temporal dependencies in the data, achieving competitive performance, as indicated by the low values of the MAE, MAPE, and RMSE metrics on the validation set.
​
Check the full code here:
​
London Bike Rides
A data analysis project: from collecting data programmatically using the Kaggle API, through data exploration, cleaning and manipulation using the pandas library in Python, to data visualization in Tableau.
​
The main goal is to create an interactive dashboard in order to visualize data on bike rides in London, between 2015-2017, showing: the total and the moving average of bike rides in a pre-selected time interval, as well as a heatmap of temperature vs. wind speed for the same period.
​
Check the full code here:
​
