Here are 6 interesting facts about Decision Trees that you may have not heard about:
1. Other than most machine learning models which are black boxes, the predictions made by decision trees are explainable.
Next facts 👇
Image: Decision trees example
1. Other than most machine learning models which are black boxes, the predictions made by decision trees are explainable.
Next facts 👇
Image: Decision trees example
2. Decision trees do not need the input features to be scaled (either normalization or standardization).
3. They can handle categorical features (Scikit-Learn doesn't support this, but TensorFlow newest implementation of trees does).
More facts 👇
3. They can handle categorical features (Scikit-Learn doesn't support this, but TensorFlow newest implementation of trees does).
More facts 👇
4. Different to most ML models that don't accept missing values, decision trees accept them.
5. They can handle imbalanced datasets(you only have to adjust the weights of the classes).
One more fact👇
5. They can handle imbalanced datasets(you only have to adjust the weights of the classes).
One more fact👇
6. Decision trees can provide the feature importances or how much each feature contributed to the model training results.
Thank you for reading this thread.
I am actively writing about Machine Learning ideas, concepts, techniques and best practices. You can support me by sharing the tweet or follow @Jeande_d.
I have more content coming and stay tuned for other things I have for my followers!
I am actively writing about Machine Learning ideas, concepts, techniques and best practices. You can support me by sharing the tweet or follow @Jeande_d.
I have more content coming and stay tuned for other things I have for my followers!
Loading suggestions...