To explain the results between training data and cross validation data after critical analysis, it can be helpful to examine the characteristics of the training data and cross validation data, as well as the model's architecture and hyperparameters. For example, if the training data and cross validation data are very similar in terms of their distribution and characteristics, and the model has a reasonable number of parameters and is not overfitting, it is likely that the model will perform well on both sets of data. On the other hand, if the training data and cross validation data are significantly different, or if the model has a large number of parameters and is prone to overfitting, the model's performance may be worse on the cross validation data compared to the training data

@Abderrahmane and Uditha, thanks for your view and contribution.