Update losses on docstrings. (#795)

keras-team · Aug 26, 2023 · add0821 · add0821
1 parent 67626f9
commit add0821
Showing 1 changed file with 6 additions and 6 deletions.
diff --git a/keras_core/metrics/accuracy_metrics.py b/keras_core/metrics/accuracy_metrics.py
@@ -48,7 +48,7 @@ class Accuracy(reduction_metrics.MeanMetricWrapper):
 
     ```python
     model.compile(optimizer='sgd',
-                  loss='mse',
+                  loss='binary_crossentropy',
                   metrics=[keras_core.metrics.Accuracy()])
     ```
     """
@@ -108,7 +108,7 @@ class BinaryAccuracy(reduction_metrics.MeanMetricWrapper):
 
     ```python
     model.compile(optimizer='sgd',
-                  loss='mse',
+                  loss='binary_crossentropy',
                   metrics=[keras_core.metrics.BinaryAccuracy()])
     ```
     """
@@ -194,7 +194,7 @@ class CategoricalAccuracy(reduction_metrics.MeanMetricWrapper):
 
     ```python
     model.compile(optimizer='sgd',
-                  loss='mse',
+                  loss='categorical_crossentropy',
                   metrics=[keras_core.metrics.CategoricalAccuracy()])
     ```
     """
@@ -278,7 +278,7 @@ class SparseCategoricalAccuracy(reduction_metrics.MeanMetricWrapper):
 
     ```python
     model.compile(optimizer='sgd',
-                  loss='mse',
+                  loss='sparse_categorical_crossentropy',
                   metrics=[keras_core.metrics.SparseCategoricalAccuracy()])
     ```
     """
@@ -349,7 +349,7 @@ class TopKCategoricalAccuracy(reduction_metrics.MeanMetricWrapper):
 
     ```python
     model.compile(optimizer='sgd',
-                  loss='mse',
+                  loss='categorical_crossentropy',
                   metrics=[keras_core.metrics.TopKCategoricalAccuracy()])
     ```
     """
@@ -423,7 +423,7 @@ class SparseTopKCategoricalAccuracy(reduction_metrics.MeanMetricWrapper):
 
     ```python
     model.compile(optimizer='sgd',
-                  loss='mse',
+                  loss='sparse_categorical_crossentropy',
                   metrics=[keras_core.metrics.SparseTopKCategoricalAccuracy()])
     ```
     """