1. Mahine Learning means learning from Data.
2. Machine = Your machine/computer Learning = Finding pattern from data.
3. Machine Learning is just Data + Algorithms, but Data is more important.
4. Feature extraction is key. If total prediction power is 100% then the effort of feature engineering = 80% and the effort of the learning algorithm = 20%.
5. Overfitting is when your algorithm is memorizing instead of learning
6. If you have small amounts of data then you're better off using more simple models (linear & logistic regression). If you have large amounts of data you can try out more complex models (Deep Learning, etc.)
7. To avoid overfitting, always use regularization
8. Training is the most important part of Machine Learning. Choose your features and hyper parameters carefully.
9. Machines don't take decisions, people do.
10. Data cleaning is the most important part of Maching Learning. You know the saying: Garbage in Garbage out.
10a. Data cleaning is a large part of #DataScience. Don't be superised if you spend more time here than you do with your friends and loved ones.
网友评论