=================================================================================

Figure 3796a shows the dependence of training score and cross-validation score on training examples. The "Training Score" or "Training Error" typically refers to the performance of a machine learning model on the training data it was trained on. It measures how well the model fits or predicts the training data. The training score is a measure of how closely the model's predictions match the actual target values in the training dataset. It is often used as a diagnostic tool during model development, but it may not provide a complete picture of the model's performance.

Figure 3796a. Dependence of training score and cross-validation score on training examples. (code)

As shown in Figure 3796b, in machine learning, more examples give your model a broader perspective and help it generalize better so that the variance is reduced. Figure 3806cb shows the effect of training data size on error. The training error and validation error can be used as proxies for bias and variance so that they can represent variance in machine learning.

Figure 3796b. Effect of training data size on error. (code)

============================================