Un poids est une valeur numérique apprise qui détermine dans quelle mesure une entrée, une caractéristique ou une connexion influence la sortie d’un modèle. Pendant l’entraînement, un modèle ajuste les poids pour réduire l’erreur et améliorer les prédictions.
A value that a model multiplies by another value. Training is the process of determining a model's ideal weights; inference is the process of using those learned weights to make predictions. Imagine a linear model with two features. Suppose that training determines the following weights (and bias): - The bias, b, has a value of 2.2 - The weight, w~1~ associated with one feature is 1.5. - The weight, w~2~ associated with the other feature is 0.4. Now imagine an example with the following feature values: - The value of one feature, x~1~, is 6. - The value of the other feature, x~2~, is 10. This linear model uses the following formula to generate a prediction, y': Therefore, the prediction is: If a weight is 0, then the corresponding feature doesn't contribute to the model. For example, if w~1~ is 0, then the value of x~1~ is irrelevant. See Linear regression in Machine Learning Crash Course for more information.