Projects are organized into branches. Change branch to access them.
Current projects:
- used cars: regression on prices of used cars based on several predictors such as mileage, year, engine size etc.
- MNIST: reaching over 97% accuracy on popular 70000 handwritten digits dataset.
- mobile prices: classification, both binary and multiclass. Each observation is a mobile, that has to be classified in a price range based on predictors such as RAM and battery power.
- wish clothes: binary classification of summer clothes in "successful" (high sales) or not (low sales).
- wine quality: classification of wines: predicting quality score.