dp-100-data-scientist-assoc video for concept: Training machine learning models varies in complexity. Simple models with small datasets can often be trained in
Concept: Training machine learning models varies in complexity. Simple models with small datasets can often be trained in a single step. Larger datasets and complex models require multiple training iterations, where the model is repeatedly adjusted using training data. Each training cycle compares the model's output with the expected label. If the prediction accuracy is sufficient, the model is considered trained. If not, incremental adjustments are made, and the process loops again until the model is optimized. During training, certain values influence how the model is adjusted. For example: Learning rate controls the extent of adjustments made in each training cycle. A high learning rate speeds up training but may prevent fine-tuning , making the model less optimal . Question: Which of the following terms best describes these values, which impact how a model is fit during iterative training?