Lint the Transfer learning with TensorFlow Hub tutorial, change Keras preprocessing APIs to stable

PiperOrigin-RevId: 402839323

Author: tensorflower-gardener

site/en/tutorials/images/transfer_learning_with_hub.ipynb

@@ -68,9 +68,9 @@
         "\n",
         "This tutorial demonstrates how to:\n",
         "\n",
-        "1. Use models from TensorFlow Hub with `tf.keras`\n",
-        "1. Use an image classification model from TensorFlow Hub\n",
-        "1. Do simple transfer learning to fine-tune a model for your own image classes"
+        "1. Use models from TensorFlow Hub with `tf.keras`.\n",
+        "1. Use an image classification model from TensorFlow Hub.\n",
+        "1. Do simple transfer learning to fine-tune a model for your own image classes."
       ]
     },
     {
@@ -333,7 +333,7 @@
         "id": "jFHdp18ccah7"
       },
       "source": [
-        "First, load this data into the model using the image data off disk with `tf.keras.preprocessing.image_dataset_from_directory`, which will generate a `tf.data.Dataset`:"
+        "First, load this data into the model using the image data off disk with `tf.keras.utils.image_dataset_from_directory`, which will generate a `tf.data.Dataset`:"
       ]
     },
     {
@@ -348,7 +348,7 @@
         "img_height = 224\n",
         "img_width = 224\n",
         "\n",
-        "train_ds = tf.keras.preprocessing.image_dataset_from_directory(\n",
+        "train_ds = tf.keras.utils.image_dataset_from_directory(\n",
         "  str(data_root),\n",
         "  validation_split=0.2,\n",
         "  subset=\"training\",\n",
@@ -394,7 +394,7 @@
         "id": "L0Btd0V3C8h4"
       },
       "source": [
-        "Second, because TensorFlow Hub's convention for image models is to expect float inputs in the `[0, 1]` range, use the `tf.keras.layers.experimental.preprocessing.Rescaling` layer to achieve this."
+        "Second, because TensorFlow Hub's convention for image models is to expect float inputs in the `[0, 1]` range, use the `tf.keras.layers.Rescaling` preprocessing layer to achieve this."
       ]
     },
     {
@@ -403,7 +403,7 @@
         "id": "Rs6gfO-ApTQW"
       },
       "source": [
-        "Note: You could also include the `tf.keras.layers.experimental.preprocessing.Rescaling` layer inside the model. Refer to the [Working with preprocessing layers](https://www.tensorflow.org/guide/keras/preprocessing_layers) guide for a discussion of the tradeoffs."
+        "Note: You could also include the `tf.keras.layers.Rescaling` layer inside the model. Refer to the [Working with preprocessing layers](https://www.tensorflow.org/guide/keras/preprocessing_layers) guide for a discussion of the tradeoffs."
       ]
     },
     {
@@ -414,7 +414,7 @@
       },
       "outputs": [],
       "source": [
-        "normalization_layer = tf.keras.layers.experimental.preprocessing.Rescaling(1./255)\n",
+        "normalization_layer = tf.keras.layers.Rescaling(1./255)\n",
         "train_ds = train_ds.map(lambda x, y: (normalization_layer(x), y)) # Where x—images, y—labels.\n",
         "val_ds = val_ds.map(lambda x, y: (normalization_layer(x), y)) # Where x—images, y—labels."
       ]