Break codeblock in sequence_models_tutorial.py into a few cells

Author: chsasank

beginner_source/nlp/sequence_models_tutorial.py

@@ -121,6 +121,8 @@
 # vector. Note this implies immediately that the dimensionality of the
 # target space of :math:`A` is :math:`|T|`.
 #
+#
+# Prepare data:
 
 def prepare_sequence(seq, to_ix):
     idxs = [to_ix[w] for w in seq]
@@ -144,6 +146,8 @@ def prepare_sequence(seq, to_ix):
 EMBEDDING_DIM = 6
 HIDDEN_DIM = 6
 
+######################################################################
+# Create the model:
 
 class LSTMTagger(nn.Module):
 
@@ -176,6 +180,9 @@ def forward(self, sentence):
         tag_scores = F.log_softmax(tag_space)
         return tag_scores
 
+######################################################################
+# Train the model:
+
 model = LSTMTagger(EMBEDDING_DIM, HIDDEN_DIM, len(word_to_ix), len(tag_to_ix))
 loss_function = nn.NLLLoss()
 optimizer = optim.SGD(model.parameters(), lr=0.1)
@@ -241,7 +248,8 @@ def forward(self, sentence):
 #
 # To get the character level representation, do an LSTM over the
 # characters of a word, and let :math:`c_w` be the final hidden state of
-# this LSTM. Hints: 
+# this LSTM. Hints:
+#
 # * There are going to be two LSTM's in your new model.
 #   The original one that outputs POS tag scores, and the new one that
 #   outputs a character-level representation of each word.