Knyati, I don't think it is correct to use R-squared as a goodness-of-fit measure. It doesn't have such inherent qualities and in fact a model which provides a good fit can produce a low R-squared value while a very poorly fitted model can result in a very high R-squared value. I've made the case against using R-squared as a goodness-of-fit measure in detail here https://www.kdnuggets.com/2020/07/r-squared-predictive-capacity-statistical-adequacy.html I think this statistic is useful as a measure of the proportion of variance explained by the predictors, but note that assumes a proper model was fitted.