How can hyper-parameters be defined within a model?

User-defined configurations are indeed how hyper-parameters are defined within a model. Hyper-parameters are settings that you configure before training a model to control the learning process and influence performance. These aspects can include learning rate, batch size, number of epochs, and other settings that are manually set by the user or data scientist to achieve the desired performance.

The hyper-parameters differ from model parameters, which are learned during the training process from the data. By customizing these configurations, practitioners can optimize their models for specific tasks or datasets, thus improving accuracy or reducing overfitting. This ability to define hyper-parameters is crucial for model tuning and effectively leveraging the capabilities of machine learning frameworks.

Other options refer to concepts that are not aligned with the definition of hyper-parameters. Default configurations imply set values that may not be tailored for specific tasks, while automatic adjustments reference adaptive learning processes not related to user-defined settings. Lastly, variables generated by the dataset refer more to the data itself rather than to the specific configurations used to control the training of a machine learning model.