Reformat code that doesn't follow the standard.

PiperOrigin-RevId: 800095985

Author: hertschuh

tf_keras/metrics/confusion_metrics.py

@@ -1624,15 +1624,15 @@ def result(self):
             # ROC or PR values. In particular
             # 1) Both measures diverge when there are no negative examples;
             # 2) Both measures diverge when there are no true positives;
-            # 3) Recall gain becomes negative when the recall is lower than the label
-            # average (i.e. when more negative examples are classified positive
-            # than real positives).
+            # 3) Recall gain becomes negative when the recall is lower than the
+            # label average (i.e. when more negative examples are classified
+            # positive than real positives).
             #
             # We ignore case 1 as it is easily communicated. For case 2 we set
             # recall_gain to 0 and precision_gain to 1. For case 3 we set the
             # recall_gain to 0. These fixes will result in an overastimation of
-            # the AUC for estimateors that are anti-correlated with the label (at
-            # some thresholds).
+            # the AUC for estimateors that are anti-correlated with the label
+            # (at some thresholds).
             #
             # The scaling factor $\frac{P}{N}$ that is used to form both
             # gain values.
@@ -1641,13 +1641,27 @@ def result(self):
                 tf.math.add(self.false_positives, self.true_negatives),
             )
 
-            recall_gain = 1.0 - scaling_factor * tf.math.divide_no_nan(self.false_negatives, self.true_positives)
-            precision_gain = 1.0 - scaling_factor * tf.math.divide_no_nan(self.false_positives, self.true_positives)
+            recall_gain = 1.0 - scaling_factor * tf.math.divide_no_nan(
+                self.false_negatives, self.true_positives
+            )
+            precision_gain = 1.0 - scaling_factor * tf.math.divide_no_nan(
+                self.false_positives, self.true_positives
+            )
             # Handle case 2.
-            recall_gain = tf.where(tf.equal(self.true_positives, 0.0), tf.zeros_like(recall_gain), recall_gain)
-            precision_gain = tf.where(tf.equal(self.true_positives, 0.0), tf.ones_like(precision_gain), precision_gain)
+            recall_gain = tf.where(
+                tf.equal(self.true_positives, 0.0),
+                tf.zeros_like(recall_gain),
+                recall_gain
+            )
+            precision_gain = tf.where(
+                tf.equal(self.true_positives, 0.0),
+                tf.ones_like(precision_gain),
+                precision_gain
+            )
             # Handle case 3.
-            recall_gain = tf.math.maximum(recall_gain, tf.zeros_like(recall_gain))
+            recall_gain = tf.math.maximum(
+                recall_gain, tf.zeros_like(recall_gain)
+            )
 
             x = recall_gain
             y = precision_gain

tf_keras/metrics/confusion_metrics_test.py

@@ -1623,7 +1623,7 @@ def test_weighted_pr_gain_majoring(self):
         # precision_gain = 1 - 7/3 [ 3/7, 0/4, 0/0 ] = [0, 1, 1]
         # heights = [max(0, 1), max(1, 1)] = [1, 1]
         # widths = [(1 - 0), (0 - 0)] = [1, 0]
-        expected_result = 1 * 1  + 0 * 1
+        expected_result = 1 * 1 + 0 * 1
         self.assertAllClose(self.evaluate(result), expected_result, 1e-3)
 
     def test_weighted_pr_gain_minoring(self):
@@ -1649,7 +1649,9 @@ def test_weighted_pr_gain_minoring(self):
 
     def test_weighted_pr_gain_interpolation(self):
         self.setup()
-        auc_obj = metrics.AUC(num_thresholds=self.num_thresholds, curve="PR_GAIN")
+        auc_obj = metrics.AUC(
+            num_thresholds=self.num_thresholds, curve="PR_GAIN"
+        )
         self.evaluate(tf.compat.v1.variables_initializer(auc_obj.variables))
         result = auc_obj(
             self.y_true, self.y_pred, sample_weight=self.sample_weight
@@ -1666,11 +1668,13 @@ def test_weighted_pr_gain_interpolation(self):
 
     def test_pr_gain_interpolation(self):
         self.setup()
-        auc_obj = metrics.AUC(num_thresholds=self.num_thresholds, curve="PR_GAIN")
+        auc_obj = metrics.AUC(
+            num_thresholds=self.num_thresholds, curve="PR_GAIN"
+        )
         self.evaluate(tf.compat.v1.variables_initializer(auc_obj.variables))
         y_true = tf.constant([0, 0, 0, 1, 0, 1, 0, 1, 1, 1])
         y_pred = tf.constant([0.1, 0.2, 0.3, 0.3, 0.4, 0.4, 0.6, 0.6, 0.8, 0.9])
-        result = auc_obj( y_true, y_pred)
+        result = auc_obj(y_true, y_pred)
 
         # tp = [5, 3, 0], fp = [5, 1, 0], fn = [0, 2, 5], tn = [0, 4, 4]
         # scaling_factor (P/N) = 5/5 = 1