Update README.md

README.md

@@ -24,12 +24,20 @@ It achieves the following results on the evaluation set:
 
 ## Model description
 
-More information needed
+This model is a GPT2-like model for generating capsid amino acid sequences. It was trained exclusively on capsid aa_seqs of Piccovirales members.
 
 ## Intended uses & limitations
 
+As a typical GPT model, it can be used to generate new sequences or used to evaluate the perplexity of given sequences.
+
 ### Generate novel sequences for viral capsid proteins
 
+```python
+from transformers import pipeline
+protgpt2 = pipeline('text-generation', model="avuhong/PiccoviralesGPT")
+sequences = protgpt2("<|endoftext|>", max_length=750, do_sample=True, top_k=950, repetition_penalty=1.2, num_return_sequences=10, eos_token_id=0)
+```
+
 ### Calculate the perplexity of a protein sequence
 
 ```python
@@ -67,7 +75,7 @@ print(f"{calculatePerplexity(seq, model, tokenizer):.2f}")
 
 ## Training and evaluation data
 
-More information needed
+Traning script is included in bash file in this repository.
 
 ## Training procedure