Merge pull request tensorflow#1 from nczempin/nczempin-patch-1

typos in jupyter notebook

courses/udacity_intro_to_tensorflow_for_deep_learning/l05c03_exercise_flowers_with_data_augmentation.ipynb

@@ -240,7 +240,7 @@
         "\n",
         "\u003cpre style=\"font-size: 10.0pt; font-family: Arial; line-height: 2; letter-spacing: 1.0pt;\" \u003e\n",
         "\u003cb\u003eflower_photos\u003c/b\u003e\n",
-        "|__ \u003cb\u003ediasy\u003c/b\u003e\n",
+        "|__ \u003cb\u003edaisy\u003c/b\u003e\n",
         "|__ \u003cb\u003edandelion\u003c/b\u003e\n",
         "|__ \u003cb\u003eroses\u003c/b\u003e\n",
         "|__ \u003cb\u003esunflowers\u003c/b\u003e\n",
@@ -255,7 +255,7 @@
         "\n",
         "\u003cpre style=\"font-size: 10.0pt; font-family: Arial; line-height: 2; letter-spacing: 1.0pt;\" \u003e\n",
         "\u003cb\u003eflower_photos\u003c/b\u003e\n",
-        "|__ \u003cb\u003ediasy\u003c/b\u003e\n",
+        "|__ \u003cb\u003edaisy\u003c/b\u003e\n",
         "|__ \u003cb\u003edandelion\u003c/b\u003e\n",
         "|__ \u003cb\u003eroses\u003c/b\u003e\n",
         "|__ \u003cb\u003esunflowers\u003c/b\u003e\n",
@@ -749,7 +749,7 @@
       "source": [
         "# TODO: Experiment with Different Parameters\n",
         "\n",
-        "So far you've created a CNN with 3 convolutional layers and followed by a fully connected layer with 512 units. In the cells below create a new CNN with a different architecture. Feel free to experiement by changing as many parameters as you like. For example, you can add more convolutional layers, or more fully connected layers. You can also experiement with different filter sizes in your convolutional layers, differnt number of units in your fully connected layers, different dropout rates, etc... You can also experiment by performing image aumentation with more image transformations that we have seen so far. Take a look at the  [ImageDataGenerator Documentation](https://keras.io/preprocessing/image/) to see a full list of all the available image transformations. For example, you can add shear transformations, or you can vary the brightness of the images, etc... Experiement as much as you can and compare the accuracy of your various models.  Which parameters give you the best result?"
+        "So far you've created a CNN with 3 convolutional layers and followed by a fully connected layer with 512 units. In the cells below create a new CNN with a different architecture. Feel free to experiement by changing as many parameters as you like. For example, you can add more convolutional layers, or more fully connected layers. You can also experiment with different filter sizes in your convolutional layers, different number of units in your fully connected layers, different dropout rates, etc... You can also experiment by performing image augmentation with more image transformations than we have seen so far. Take a look at the  [ImageDataGenerator Documentation](https://keras.io/preprocessing/image/) to see a full list of all the available image transformations. For example, you can add shear transformations, or you can vary the brightness of the images, etc... Experiment as much as you can and compare the accuracy of your various models.  Which parameters give you the best results?"
       ]
     },
     {