Update README.md

README.md

@@ -7,22 +7,26 @@ This is a paraphrasing model for the Persian language. It is based on [the monol
 ```python
 
 >>> pip install transformers
->>> from transformers import (T5ForConditionalGeneration, AutoTokenizer)
+>>> from transformers import (T5ForConditionalGeneration, AutoTokenizer, pipeline)
 >>> import torch
 
 model_path = 'erfan226/persian-t5-paraphraser'
 model = T5ForConditionalGeneration.from_pretrained(model_path)
 tokenizer = AutoTokenizer.from_pretrained(model_path)
+pipe = pipeline(task='text2text-generation', model=model, tokenizer=tokenizer)
 
 def paraphrase(text):
-    input = tokenizer(text, return_tensors='pt', padding=True).to(model.device)
-    max_size = int(input.input_ids.shape[1] * 1.5 + 10)
-    out = model.generate(**input, encoder_no_repeat_ngram_size=4, do_sample=False, num_beams=10, max_length=max_size, no_repeat_ngram_size=4,)
-    return tokenizer.decode(out[0], skip_special_tokens=True)
+  for j in range(5):
+    out = pipe(text, encoder_no_repeat_ngram_size=5, do_sample=True, num_beams=5, max_length=128)[0]['generated_text']
+    print("Paraphrase:", out)
 
-for text1, text2 in zip(x, y):
-    print("Original:", text1)
-    print("Paraphrase:", paraphrase(text1))
-    print("Original Paraphrase:", text2)
+text = "این یک مقالهٔ خرد آلمان است. می‌توانید با گسترش آن به ویکی‌پدیا کمک کنید."
+print("Original:", text)
+paraphrase(text)
 
-```
+```
+
+# Training data
+
+```python
+This model was trained on the Persian subset of the [Tapaco dataset](https://huggingface.co/datasets/tapaco).