Model training and evaluation
The process of training and validating or testing a model consists of the following three major steps if you use scikit-learn or PyTorch and TensorFlow for neural network modeling:
- Initializing the model: Initializing a model is about specifying the method, its hyperparameters, and the random state to be used for modeling.
- Training the model: In model training, the initialized model in Step 1 gets used on the training data to train a machine learning model.
- Inference, assignment, and performance assessment: In this step, the trained model can be used for inference (for example, predicting outputs) in supervised learning or, for example, assigning new data points to identified clusters in unsupervised learning. In supervised learning, you can use these predictions for model performance assessment.
These steps are similar for both supervised learning and unsupervised learning models. In Steps 1 and 2, both types of models can be trained...
