Features are the input variables used to make predictions. For example, house size, number of bedrooms, and location when predicting house price.
Labels are the expected outputs the model is trained to predict. In the house price example, the label is the actual price.
Training is the process of feeding the model data so it can learn patterns. You provide both features and labels.
Testing evaluates the model on unseen data to check its performance and generalization.
The model is the algorithm plus learned parameters. It’s the trained system that makes predictions.
Prediction is using the trained model to estimate outputs for new inputs.
| Concept | Role in ML Workflow | Code Representation |
|---|---|---|
| Features | Input variables | X |
| Labels | Expected output | y |
| Training | Learning from known data | model.fit(X, y) |
| Testing | Evaluating on new data | model.predict(X) |
| Model | The trained algorithm | model |
| Prediction | Output from the model | y_pred |