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?

  • Search grid enter image description here

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.