Fix in documentation of "Image classification from scratch" example.

Fixes #2199

Author: hertschuh

examples/vision/image_classification_from_scratch.py

@@ -187,8 +187,7 @@ def data_augmentation(images):
 augmented images, like this:
 
 ```python
-augmented_train_ds = train_ds.map(
-    lambda x, y: (data_augmentation(x, training=True), y))
+augmented_train_ds = train_ds.map(lambda x, y: (data_augmentation(x), y))
 ```
 
 With this option, your data augmentation will happen **on CPU**, asynchronously, and will

examples/vision/ipynb/image_classification_from_scratch.ipynb

@@ -324,8 +324,7 @@
     "augmented images, like this:\n",
     "\n",
     "```python\n",
-    "augmented_train_ds = train_ds.map(\n",
-    "    lambda x, y: (data_augmentation(x, training=True), y))\n",
+    "augmented_train_ds = train_ds.map(lambda x, y: (data_augmentation(x), y))\n",
     "```\n",
     "\n",
     "With this option, your data augmentation will happen **on CPU**, asynchronously, and will\n",

examples/vision/md/image_classification_from_scratch.md

@@ -236,8 +236,7 @@ If you're training on GPU, this may be a good option.
 augmented images, like this:
 
 ```python
-augmented_train_ds = train_ds.map(
-    lambda x, y: (data_augmentation(x, training=True), y))
+augmented_train_ds = train_ds.map(lambda x, y: (data_augmentation(x), y))
 ```
 
 With this option, your data augmentation will happen **on CPU**, asynchronously, and will