reflected feedback

notebooks/multi_modal/solutions/image_captioning.ipynb

@@ -111,9 +111,12 @@
     "of the images in the original dataset did not have captions)\n",
     "\n",
     "First, let's define some constants.<br>\n",
-    "In this lab, we will use a pretrained [InceptionResNetV2](https://www.tensorflow.org/api_docs/python/tf/keras/applications/inception_resnet_v2/InceptionResNetV2) model as a feature extractor, so some constants are comming from the InceptionResNetV2 model definition.\n",
+    "In this lab, we will use a pretrained [InceptionResNetV2](https://www.tensorflow.org/api_docs/python/tf/keras/applications/inception_resnet_v2/InceptionResNetV2) model from `tf.keras.applications` as a feature extractor, so some constants are comming from the InceptionResNetV2 model definition.<br>\n",
+    "So if you want to use other type of base model, please make sure to change these constants as well.\n",
     "\n",
-    "So if you want to use other type of base model, please make sure to change these constans as well."
+    "`tf.keras.applications` is a pretrained model repository like [TensorFlow Hub](https://tfhub.dev), but while Tensorflow Hub hosts models for different modalities including image, text, audio, and so on, `tf.keras.application` only hosts popular and stable models for images.<br>\n",
+    "However, `tf.keras.applications` is more flexible as it contains model metadata and it allow us to access and control the model behavior, while most of the TensorFlow Hub based models that only contains compiled SavedModels.<br>\n",
+    "So, for example, we can get output not only from the final layer of the model (e.g. flattend 1D Tensor output of CNN models), but also from intermediate layers (e.g. intermediate 3D Tensor) by accessing layer metadata."
    ]
   },
   {
@@ -166,8 +169,8 @@
     "- rescale pixel values to [0,1]\n",
     "- return image(`image_tensor`) and captions(`captions`) dictionary.\n",
     "\n",
-    "**Note**: This dataset is too large to store in an ephemeral location. Therefore, It is stored in a public GCS bucket. \n",
-    "If you access it from a Notebook outside the US, it will be (a) slow and (b) subject to a network charge."
+    "**Note**: This dataset is too large to store in an local environment. Therefore, It is stored in a public GCS bucket located in us-central1. \n",
+    "So if you access it from a Notebook outside the US, it will be (a) slow and (b) subject to a network charge."
    ]
   },
   {