An article written by Jay Alammar (http://jalammar.github.io/a-visual-guide-to-using-bert-for-the-first-time/) on using a BERT transformer for text classification.
The article mentions the following use of grid search to find the best fit value of C to increase the models' regularization strength for better training. The C score can then be applied to a LogisticRegression model to fine-tune the model's accuracy better.
If the label used changed to numbers between 0 and 1, e.g., 0.01, 0.00043, 0.0567, a LinearRegression model fits the continuous data better than the LogisticRegression model.
Is there the same way to find the best fit value for a LinearRegression model, similar to the grid search method used to better fine-tune a text classification model?