cristianbmj / portfoliodatascience

Repository of projects Data Science

Projects of Data Science

Machine Learning

Stars Categorization: Proyect classification's Machine Learning of Start Giants or Dwarfs , consider the dataset. The categorization the Stars was studied according to criteria the Morgan–Keenan (MK) classification system. Firstly, the dataset was cleaned and then it was balanced. In order to apply three models's Machine Learning, being Random Forest Classification the one with the better results, obtaining a precision of 93%.

Forecasting Store Sales: Forecasting to determine Store Sales of Ecuador. The Stores are in locations in the country. The Model learning is the kind Hydrid, it be first model RandomForestRegressor while the second model is LinearRegression.(under build-up)

Deep Learning

Petal of the Metal: The present Machine learning model, consists to classify 104 types of flowers based on their images. For that, added news layers were implemented and transfer learning used pretrained apliccations Xception and DenseNet201 to get a Hybric Model. The train it on a Tensor Processing Unit (TPU).