Adds usage example.

README.md

@@ -3,3 +3,20 @@ This is a small cross attention entailment model trained for zero-shot and few-s
 The base model is [mpnet-base](https://huggingface.co/microsoft/mpnet-base) and it has been trained with the code from [here](https://github.com/facebookresearch/anli).
 
 The model has been trained using [SNLI](https://nlp.stanford.edu/projects/snli/) and [MNLI](https://cims.nyu.edu/~sbowman/multinli/).
+
+Usage:
+
+```python
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
+import torch
+import numpy as np
+
+model = AutoModelForSequenceClassification.from_pretrained("symanto/mpnet-base-snli-mnli")
+tokenizer = AutoTokenizer.from_pretrained("symanto/mpnet-base-snli-mnli")
+
+inputs = tokenizer(["I like this pizza. [SEP] The sentence is positive.",  "I like this pizza. [SEP] The sentence is negative."], return_tensors="pt")
+logits = model(**inputs).logits
+probs =  torch.softmax(logits, dim=1).tolist()
+print("probs", probs)
+np.testing.assert_almost_equal(probs, [[0.66, 0.33, 0.01], [0.08, 0.22, 0.70]], decimal=2)
+```