Merge pull request #13 from grant/patch-1

Update README.md

README.md

@@ -30,66 +30,69 @@ The complete documentation for PyTorch-NLP is available via [our ReadTheDocs web
 
 Add PyTorch-NLP to your project by following one the common use cases:
 
-- From the [neural network package](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.nn.html),
-  use a Simple Recurrent Unit (SRU), like so:
+### From the [neural network package](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.nn.html), use a Simple Recurrent Unit (SRU)
 
-    ```python
-    from torchnlp.nn import SRU
-    import torch
+```python
+from torchnlp.nn import SRU
+import torch
 
-    input_ = torch.autograd.Variable(torch.randn(6, 3, 10))
-    sru = SRU(10, 20)
+input_ = torch.autograd.Variable(torch.randn(6, 3, 10))
+sru = SRU(10, 20)
 
-    # Apply a Simple Recurrent Unit to `input_`
-    sru(input_) # RETURNS: (output [torch.FloatTensor of size 6x3x20], hidden_state [torch.FloatTensor of size 2x3x20])
-    ```
+# Apply a Simple Recurrent Unit to `input_`
+sru(input_)
+# RETURNS: (
+#   output [torch.FloatTensor (6x3x20)], 
+#   hidden_state [torch.FloatTensor (2x3x20)]
+# )
+```
 
-- Load a [dataset](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.datasets.html) like IMDB.
+### Load a [dataset](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.datasets.html) like IMDB
 
-    ```python
-    from torchnlp.datasets import imdb_dataset
-
-    # Load the imdb training dataset
-    train = imdb_dataset(train=True)
-    train[0]  # RETURNS: {'text': 'For a movie that gets..', 'sentiment': 'pos'}
-    ```
-      
-- Encode text into vectors with the [text encoders package](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.text_encoders.html).
-
-    ```python
-    from torchnlp.text_encoders import WhitespaceEncoder
-
-    # Create a `WhitespaceEncoder` with a corpus of text
-    encoder = WhitespaceEncoder(["now this ain't funny", "so don't you dare laugh"])
-
-    # Encode and decode phrases
-    encoder.encode("this ain't funny.") # RETURNS: torch.LongTensor([6, 7, 1])
-    encoder.decode(encoder.encode("This ain't funny.")) # RETURNS: "this ain't funny."
-    ```
-
-- Load FastText, state-of-the-art English [word vector representations](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.word_to_vector.html).
+```python
+from torchnlp.datasets import imdb_dataset
 
-    ```python
-    from torchnlp.word_to_vector import FastText
-
-    vectors = FastText()
-    # Load vectors for any word as a `torch.FloatTensor`
-    vectors['hello']  # RETURNS: [torch.FloatTensor of size 100]
-    ```
-
-- Compute the BLEU Score with the [metrics package](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.metrics.html).
+# Load the imdb training dataset
+train = imdb_dataset(train=True)
+train[0]  # RETURNS: {'text': 'For a movie that gets..', 'sentiment': 'pos'}
+```
 
-    ```python
-    from torchnlp.metrics import get_moses_multi_bleu
-
-    hypotheses = ["The brown fox jumps over the dog 笑"]
-    references = ["The quick brown fox jumps over the lazy dog 笑"]
+### Encode text into vectors with the [text encoders package](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.text_encoders.html)
+
+```python
+from torchnlp.text_encoders import WhitespaceEncoder
+
+# Create a `WhitespaceEncoder` with a corpus of text
+encoder = WhitespaceEncoder(["now this ain't funny", "so don't you dare laugh"])
+
+# Encode and decode phrases
+encoder.encode("this ain't funny.") # RETURNS: torch.LongTensor([6, 7, 1])
+encoder.decode(encoder.encode("This ain't funny.")) # RETURNS: "this ain't funny."
+```
 
-    # Compute BLEU score with the official BLEU perl script
-    get_moses_multi_bleu(hypotheses, references, lowercase=True)  # RETURNS: 47.9
-    ```
+### Load FastText, state-of-the-art English [word vector representations](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.word_to_vector.html)
+
+```python
+from torchnlp.word_to_vector import FastText
+
+vectors = FastText()
+# Load vectors for any word as a `torch.FloatTensor`
+vectors['hello']  # RETURNS: [torch.FloatTensor of size 100]
+```
 
-Find longer examples, [here](https://github.com/PetrochukM/PyTorch-NLP/tree/master/examples).
+### Compute BLEU scores with the [metrics package](http://pytorchnlp.readthedocs.io/en/latest/source/torchnlp.metrics.html)
+
+```python
+from torchnlp.metrics import get_moses_multi_bleu
+
+hypotheses = ["The brown fox jumps over the dog 笑"]
+references = ["The quick brown fox jumps over the lazy dog 笑"]
+
+# Compute BLEU score with the official BLEU perl script
+get_moses_multi_bleu(hypotheses, references, lowercase=True)  # RETURNS: 47.9
+```
+
+Longer samples are under [`examples/`](examples/).
 
 ## Contributing
 
@@ -99,7 +102,6 @@ We've released PyTorch-NLP because we found a lack of basic toolkits for NLP in
 
 Read our [contributing guide](https://github.com/PetrochukM/PyTorch-NLP/blob/master/Contributing.md) to learn about our development process, how to propose bugfixes and improvements, and how to build and test your changes to PyTorch-NLP.
 
-
 ## Related Work
 
 ### [torchtext](https://github.com/pytorch/text)