Use dictionaries for both loss and loss_weights consistently (ker…

…as-team#500) * Use dictionaries for both `loss` and `loss_weights` consistently * Update the description text * `python autogen.py add_guide functional_api.py`
ZackPashkin · Jun 16, 2021 · b48a9a9 · b48a9a9
1 parent 51f6470
commit b48a9a9
Show file tree

Hide file tree

Showing 3 changed files with 6 additions and 6 deletions.
diff --git a/guides/functional_api.py b/guides/functional_api.py
@@ -381,7 +381,7 @@ def get_model():
 
 """
 Since the output layers have different names, you could also specify
-the loss like this:
+the losses and loss weights with the corresponding layer names:
 """
 
 model.compile(
@@ -390,7 +390,7 @@ def get_model():
         "priority": keras.losses.BinaryCrossentropy(from_logits=True),
         "department": keras.losses.CategoricalCrossentropy(from_logits=True),
     },
-    loss_weights=[1.0, 0.2],
+    loss_weights={"priority": 1.0, "department": 0.2},
 )
 
 """

diff --git a/guides/ipynb/functional_api.ipynb b/guides/ipynb/functional_api.ipynb
@@ -681,7 +681,7 @@
    },
    "source": [
     "Since the output layers have different names, you could also specify\n",
-    "the loss like this:"
+    "the losses and loss weights with the corresponding layer names:"
    ]
   },
   {
@@ -698,7 +698,7 @@
     "        \"priority\": keras.losses.BinaryCrossentropy(from_logits=True),\n",
     "        \"department\": keras.losses.CategoricalCrossentropy(from_logits=True),\n",
     "    },\n",
-    "    loss_weights=[1.0, 0.2],\n",
+    "    loss_weights={\"priority\": 1.0, \"department\": 0.2},\n",
     ")"
    ]
   },

diff --git a/guides/md/functional_api.md b/guides/md/functional_api.md
@@ -605,7 +605,7 @@ model.compile(
 ```
 
 Since the output layers have different names, you could also specify
-the loss like this:
+the losses and loss weights with the corresponding layer names:
 
 
 ```python
@@ -615,7 +615,7 @@ model.compile(
         "priority": keras.losses.BinaryCrossentropy(from_logits=True),
         "department": keras.losses.CategoricalCrossentropy(from_logits=True),
     },
-    loss_weights=[1.0, 0.2],
+    loss_weights={"priority": 1.0, "department": 0.2},
 )
 ```