nkululeko 0.88.0__tar.gz → 0.88.1__tar.gz

{nkululeko-0.88.0 → nkululeko-0.88.1}/CHANGELOG.md

@@ -1,6 +1,10 @@
 Changelog
 =========
 
+Version 0.88.1
+--------------
+* added obligatory scatter plot for regression
+
 Version 0.88.0
 --------------
 * added ensemble late fusion and AST features

{nkululeko-0.88.0/nkululeko.egg-info → nkululeko-0.88.1}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: nkululeko
-Version: 0.88.0
+Version: 0.88.1
 Summary: Machine learning audio prediction experiments based on templates
 Home-page: https://github.com/felixbur/nkululeko
 Author: Felix Burkhardt
@@ -201,7 +201,7 @@ Here is an overview of the interfaces/modules:
 All of them take *--config <my_config.ini>* as an argument.
 
 * **nkululeko.nkululeko**: do machine learning experiments combining features and learners
-* **nkululeko.ensemble**: combine several nkululeko experiments and report on late fusion results
+* **nkululeko.ensemble**: [combine several nkululeko experiments](http://blog.syntheticspeech.de/2024/06/25/nkululeko-ensemble-classifiers-with-late-fusion/) and report on late fusion results
   * *configurations*: which experiments to combine
   * *--method* (optional): majority_voting, mean, max, sum
   * *--outfile* (optional): name of CSV file for output
@@ -356,6 +356,10 @@ F. Burkhardt, Johannes Wagner, Hagen Wierstorf, Florian Eyben and Björn Schulle
 Changelog
 =========
 
+Version 0.88.1
+--------------
+* added obligatory scatter plot for regression
+
 Version 0.88.0
 --------------
 * added ensemble late fusion and AST features

{nkululeko-0.88.0 → nkululeko-0.88.1}/README.md

@@ -157,7 +157,7 @@ Here is an overview of the interfaces/modules:
 All of them take *--config <my_config.ini>* as an argument.
 
 * **nkululeko.nkululeko**: do machine learning experiments combining features and learners
-* **nkululeko.ensemble**: combine several nkululeko experiments and report on late fusion results
+* **nkululeko.ensemble**: [combine several nkululeko experiments](http://blog.syntheticspeech.de/2024/06/25/nkululeko-ensemble-classifiers-with-late-fusion/) and report on late fusion results
   * *configurations*: which experiments to combine
   * *--method* (optional): majority_voting, mean, max, sum
   * *--outfile* (optional): name of CSV file for output

{nkululeko-0.88.0 → nkululeko-0.88.1}/nkululeko/constants.py

@@ -1,2 +1,2 @@
-VERSION="0.88.0"
+VERSION="0.88.1"
 SAMPLING_RATE = 16000

{nkululeko-0.88.0 → nkululeko-0.88.1}/nkululeko/reporting/reporter.py

@@ -186,6 +186,9 @@ class Reporter:
             epoch (int, optional): Number of epoch. Defaults to None.
         """
         if not self.util.exp_is_classification():
+            self._plot_scatter(
+                self.truths, self.preds, plot_name.replace("cnf", "scatter"), epoch
+            )
             self.continuous_to_categorical()
         self._plot_confmat(self.truths, self.preds, plot_name, epoch)
 
@@ -218,6 +221,43 @@ class Reporter:
             pred = np.digitize(pred, bins) - 1
         self._plot_confmat(truth, pred.astype("int"), plot_name, 0)
 
+    def _plot_scatter(self, truths, preds, plot_name, epoch=None):
+        # print(truths)
+        # print(preds)
+        if epoch is None:
+            epoch = self.epoch
+        fig_dir = self.util.get_path("fig_dir")
+        fig = plt.figure()  # figsize=[5, 5]
+
+        pcc = pearsonr(self.truths, self.preds)[0]
+
+        reg_res = f"{self.result.test:.3f} {self.MEASURE}"
+
+        plt.scatter(truths, preds, cmap="Blues")
+        plt.xlabel("truth")
+        plt.ylabel("prediction")
+
+        if epoch != 0:
+            plt.title(f"Scatter plot: {reg_res}, PCC: {pcc:.3f}, Epoch: {epoch}")
+        else:
+            plt.title(f"Scatter plot: {reg_res}, PCC: {pcc:.3f}")
+
+        img_path = f"{fig_dir}{plot_name}{self.filenameadd}.{self.format}"
+        plt.savefig(img_path)
+        self.util.debug(f"Saved scatter plot to {img_path}")
+        fig.clear()
+        plt.close(fig)
+        plt.close()
+        plt.clf()
+        glob_conf.report.add_item(
+            ReportItem(
+                Header.HEADER_RESULTS,
+                self.util.get_model_description(),
+                "Result scatter plot",
+                img_path,
+            )
+        )
+
     def _plot_confmat(self, truths, preds, plot_name, epoch=None):
         # print(truths)
         # print(preds)

{nkululeko-0.88.0 → nkululeko-0.88.1/nkululeko.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: nkululeko
-Version: 0.88.0
+Version: 0.88.1
 Summary: Machine learning audio prediction experiments based on templates
 Home-page: https://github.com/felixbur/nkululeko
 Author: Felix Burkhardt
@@ -201,7 +201,7 @@ Here is an overview of the interfaces/modules:
 All of them take *--config <my_config.ini>* as an argument.
 
 * **nkululeko.nkululeko**: do machine learning experiments combining features and learners
-* **nkululeko.ensemble**: combine several nkululeko experiments and report on late fusion results
+* **nkululeko.ensemble**: [combine several nkululeko experiments](http://blog.syntheticspeech.de/2024/06/25/nkululeko-ensemble-classifiers-with-late-fusion/) and report on late fusion results
   * *configurations*: which experiments to combine
   * *--method* (optional): majority_voting, mean, max, sum
   * *--outfile* (optional): name of CSV file for output
@@ -356,6 +356,10 @@ F. Burkhardt, Johannes Wagner, Hagen Wierstorf, Florian Eyben and Björn Schulle
 Changelog
 =========
 
+Version 0.88.1
+--------------
+* added obligatory scatter plot for regression
+
 Version 0.88.0
 --------------
 * added ensemble late fusion and AST features