Fix typos in the IMDB tutorial.

Addresses tensorflow#4. PiperOrigin-RevId: 267654707

Author: arjung

g3doc/tutorials/graph_keras_lstm_imdb.ipynb

@@ -630,8 +630,8 @@
  "will include original sample features as well as features of their corresponding\n",
  "neighbors.\n",
  "\n",
- "In this tutorial, we consider undirected edges and we use a maximum of 1\n",
- "neighbor per sample."
+ "In this tutorial, we consider undirected edges and use a maximum of 3 neighbors\n",
+ "per sample to augment training data with graph neighbors."
  ]
  },
  {
@@ -659,7 +659,10 @@
  "source": [
  "## Base model\n",
  "\n",
- "We are now ready to build a base model without graph regularization. In order to build this model, we can either use embeddings that were used in building the graph, or we can learn new embeddings jointly along with the classification task. For the purpose of this notebook, we will do the latter."
+ "We are now ready to build a base model without graph regularization. In order to\n",
+ "build this model, we can either use embeddings that were used in building the\n",
+ "graph, or we can learn new embeddings jointly along with the classification\n",
+ "task. For the purpose of this notebook, we will do the latter."
  ]
  },
  {
@@ -701,9 +704,11 @@
  "\n",
  "- **num_classes**: There are 2 classes -- *positive* and *negative*.\n",
  "\n",
- "- **max_seq_length**: This is the maximum number of words considered from each movie review in this example.\n",
+ "- **max_seq_length**: This is the maximum number of words considered from each\n",
+ " movie review in this example.\n",
  "\n",
- "- **vocab_size**: This is the size of the vocabulary considered for this example.\n",
+ "- **vocab_size**: This is the size of the vocabulary considered for this\n",
+ " example.\n",
  "\n",
  "- **distance_type**: This is the distance metric used to regularize the sample\n",
  " with its neighbors.\n",
@@ -712,8 +717,11 @@
  " the graph regularization term in the overall loss function.\n",
  "\n",
  "- **num_neighbors**: The number of neighbors used for graph regularization.\n",
+ " This value has to be less than or equal to the `--max_nbrs` command line\n",
+ " value used above when you ran the `pack_nbrs` utility.\n",
  "\n",
- "- **num_fc_units**: The number of units in the fully connected layer of the neural network.\n",
+ "- **num_fc_units**: The number of units in the fully connected layer of the\n",
+ " neural network.\n",
  "\n",
  "- **train_epochs**: The number of training epochs.\n",
  "\n",