AWS Certified AI PractitionerAI/ML and Generative AI FundamentalsMedium
A satellite imaging company uses an AI/ML model to classify land use from aerial photographs (e.g., forest, urban, agricultural). Each photograph is labeled with one of these categories. Once the model is trained, it processes new, unlabeled photographs to assign them to a land-use category. In the context of the machine learning lifecycle, what is this final step of assigning categories to new data called?
- ATraining
- BInference
- CValidation
- DFeature Engineering
Show answer & explanationAnswer & explanation
Correct answer: B. Inference
Inference is the process of using a trained machine learning model to make predictions or classify new, unseen data. In this scenario, after the model is trained, processing new photographs to assign categories is the inference step.
Why the other options are wrong
- A. Training is the process of fitting the model to the labeled dataset, where the model learns patterns from the data.
- C. Validation is used during training to tune hyperparameters and assess model performance on unseen data before final deployment, preventing overfitting.
- D. Feature Engineering is the process of creating new features or transforming existing ones from raw data to improve model performance, occurring before training.
Inference (AI/ML Lifecycle)
The process of using a trained machine learning model to make predictions or classifications on new, unseen data.
- Occurs after a model has been trained and validated.
- Applies the learned patterns to real-world data.
- The primary goal of deploying an ML model.
- Can be performed in batch or real-time.
Memory trick: ML cycle: data, build, test, use.