comparison base_model_trainer.py @ 7:f4cb41f458fd draft default tip

planemo upload for repository https://github.com/goeckslab/gleam commit b430f8b466655878c3bf63b053655fdbf039ddb0

6:a32ff7201629

        if (
            hasattr(self, "cross_validation")
            and self.cross_validation is not None
            and self.cross_validation is False
        ):
            self.setup_params["cross_validation"] = self.cross_validation

        if hasattr(self, "cross_validation") and self.cross_validation is not None:
            if hasattr(self, "cross_validation_folds"):
                self.setup_params["fold"] = self.cross_validation_folds

        if hasattr(self, "remove_outliers") and self.remove_outliers is not None:
            self.setup_params["remove_outliers"] = self.remove_outliers

                score_func=average_precision_score,
                average="weighted",
            )

        if hasattr(self, "models") and self.models is not None:
            self.best_model = self.exp.compare_models(include=self.models)
        else:
            self.best_model = self.exp.compare_models()
        self.results = self.exp.pull()
        if self.task_type == "classification":
            self.results.rename(columns={"AUC": "ROC-AUC"}, inplace=True)

        _ = self.exp.predict_model(self.best_model)
        self.test_result_df = self.exp.pull()

                "Explainer Plots</div>"
            )
        html_content += (
            "</div>"
            '<div id="summary" class="tab-content">'
            "<h2>Model Metrics from Cross-Validation Set</h2>"
            f"<h2>Best Model: {model_name}</h2>"
            "<h5>The best model is selected by: Accuracy (Classification)"
            " or R2 (Regression).</h5>"
            f"{self.results.to_html(index=False, classes='table sortable')}"
            "<h2>Best Model's Hyperparameters</h2>"

7:f4cb41f458fd

        if (
            hasattr(self, "cross_validation")
            and self.cross_validation is not None
            and self.cross_validation is False
        ):
            logging.info(
                "cross_validation is set to False. This will disable cross-validation."
            )

        if hasattr(self, "cross_validation") and self.cross_validation:
            if hasattr(self, "cross_validation_folds"):
                self.setup_params["fold"] = self.cross_validation_folds

        if hasattr(self, "remove_outliers") and self.remove_outliers is not None:
            self.setup_params["remove_outliers"] = self.remove_outliers

                score_func=average_precision_score,
                average="weighted",
            )

        if hasattr(self, "models") and self.models is not None:
            self.best_model = self.exp.compare_models(include=self.models, cross_validation=self.cross_validation)
        else:
            self.best_model = self.exp.compare_models(cross_validation=self.cross_validation)
        self.results = self.exp.pull()

        if self.task_type == "classification":
            self.results.rename(columns={"AUC": "ROC-AUC"}, inplace=True)

        _ = self.exp.predict_model(self.best_model)
        self.test_result_df = self.exp.pull()

                "Explainer Plots</div>"
            )
        html_content += (
            "</div>"
            '<div id="summary" class="tab-content">'
            f"<h2>Model Metrics from {'Cross-Validation Set' if self.cross_validation else 'Validation set'}</h2>"
            f"<h2>Best Model: {model_name}</h2>"
            "<h5>The best model is selected by: Accuracy (Classification)"
            " or R2 (Regression).</h5>"
            f"{self.results.to_html(index=False, classes='table sortable')}"
            "<h2>Best Model's Hyperparameters</h2>"