Last active
September 23, 2020 16:31
-
-
Save lgallen/55449ee8f79836eb902a4945d29a8108 to your computer and use it in GitHub Desktop.
Shabnam project homework help
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # Grabbing the preprocessor | |
| pre = fit_model.named_steps['preprocessor'] | |
| # Getting the numerical and categorical features from the pipeline | |
| num_feats = pre.transformers_[0][2] | |
| cat_feats = pre.transformers_[1][1]['onehot']\ | |
| .get_feature_names(categorical_features) | |
| all_feats = num_feats+list(cat_feats) | |
| # Dataframe for visual examination of coefficients | |
| df_coefs = pd.DataFrame() | |
| df_coefs['feature'] = all_feats | |
| df_coefs['coefficient'] = model.coef_ | |
| # Filter out all but the coefficients with some significance | |
| df_coefs[abs(df_coefs['coefficient']) > .01].sort_values('coefficient') |
Updating to include sort_values, typo in original that excluded negative coef.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Run this after the you have trained the model in the homework submission you shared. This will help you understand what the model is doing. I think what you have now is OK, but you probably have some multicollinearity, especially among those geographic features.