Update PyTorch example in README.md

The previous pytorch example was incomplete (did not show how to process the outputs) and was also not using the specific question answering version of distilbert.

README.md

@@ -64,10 +64,10 @@ Answer: 'SQuAD dataset', score: 0.4704, start: 147, end: 160
 Here is how to use this model in PyTorch:
 
 ```python
-from transformers import DistilBertTokenizer, DistilBertModel
+from transformers import DistilBertTokenizer, DistilBertForQuestionAnswering
 import torch
 tokenizer = DistilBertTokenizer.from_pretrained('distilbert-base-uncased-distilled-squad')
-model = DistilBertModel.from_pretrained('distilbert-base-uncased-distilled-squad')
+model = DistilBertForQuestionAnswering.from_pretrained('distilbert-base-uncased-distilled-squad')
 
 question, text = "Who was Jim Henson?", "Jim Henson was a nice puppet"
 
@@ -75,7 +75,11 @@ inputs = tokenizer(question, text, return_tensors="pt")
 with torch.no_grad():
     outputs = model(**inputs)
 
-print(outputs)
+answer_start_index = torch.argmax(outputs.start_logits)
+answer_end_index = torch.argmax(outputs.end_logits)
+
+predict_answer_tokens = inputs.input_ids[0, answer_start_index : answer_end_index + 1]
+tokenizer.decode(predict_answer_tokens)
 ```
 
 And in TensorFlow: