When life gives you lemons but you'll never want a lemon car! Let's prevent that with the help of data science!
-Compare multiple machine learning algorithms.
Measured using Mean Squared Error and R2 Score
Results so far:
Random Forest Classifier:
MSE: 0.12445763873
R2: -0.153979172038
Gradient Boosting Classifier:
MSE: 0.122402374971
R2: -0.134922635252
Ada Boost Classifier:
MSE: 0.123087462891
R2: -0.141274814181
Extreme Gradient Bossting Classifier:
MSE: 0.122744918931
R2: -0.138098724717
-Data cleaning and null removal.
- Replaced null numerical values with the mean of the column.
- Replaced null string values with the most common value of the column.
- Data scaled using Robust Scaler in sklearn.
-Determine whether a feature is important in recognizing lemons(Soon)
-Use Deep Learning approach for prediction(Soon)
Include more features such as Model and TRIM