Update README.md

README.md

@@ -29,4 +29,49 @@ widget:
       کود گوگرد بنتونیتی تا پنج سال آینده به ۱۰۰ هزار تن در سال برسد.	
     example_title: Example 1
 
----
+---
+
+
+# mT5 Persian Summary
+
+This model is fine-tuned to generate summaries based on the input provided. It has been fine-tuned on a wide range of Persian news data, including [BBC news](https://huggingface.co/datasets/csebuetnlp/xlsum) and [pn_summary](https://huggingface.co/datasets/pn_summary).
+
+## Usage
+
+```
+from transformers import  AutoModelForSeq2SeqLM, MT5Tokenizer
+
+model = AutoModelForSeq2SeqLM.from_pretrained('nafisehNik/mt5-persian-summary')
+
+tokenizer = MT5Tokenizer.from_pretrained("nafisehNik/mt5-persian-summary")
+
+
+# method for summary generation, using the global model and tokenizer
+def generate_summary(model, abstract, num_beams = 2, repetition_penalty = 1.0,
+                    length_penalty = 2.0, early_stopping = True, max_output_length = 120):
+    source_encoding=tokenizer(abstract, max_length=1000, padding="max_length", truncation=True, return_attention_mask=True, add_special_tokens=True, return_tensors="pt")
+
+    generated_ids=model.generate(
+        input_ids=source_encoding["input_ids"],
+        attention_mask=source_encoding["attention_mask"],
+        num_beams=num_beams,
+        max_length=max_output_length,
+        repetition_penalty=repetition_penalty,
+        length_penalty=length_penalty,
+        early_stopping=early_stopping,
+        use_cache=True
+        )
+
+    preds=[tokenizer.decode(gen_id, skip_special_tokens=True, clean_up_tokenization_spaces=True) 
+         for gen_id in generated_ids]
+
+    return "".join(preds)
+
+text = "YOUR INPUT TEXT"
+result = generate_summary(model=model, abstract=text, num_beams=2, max_output_length=120)
+```
+
+
+## Citation
+
+If you find this model useful, make a link to the huggingface model.