Upload README.md

README.md

@@ -0,0 +1,34 @@
+---
+language: protein
+tags:
+- protein
+datasets:
+- uniref-100
+---
+
+# RITA-S
+
+RITA is a family of autoregressive protein models, developed in collaboration between Lighton, Harvard and Oxford.
+
+
+
+Model | #Params | d_model | layers | lm loss uniref-100
+--- | --- | --- | --- | --- | 
+[**Small**](https://huggingface.co/lightonai/RITA_s) | 85M  | 768 | 12 | 2.31
+[Medium](https://huggingface.co/lightonai/RITA_m) | 300M | 1024 | 24 | 2.01
+[Large](https://huggingface.co/lightonai/RITA_l)| 680M | 1536 | 24 | 1.82
+[XLarge](https://huggingface.co/lightonai/RITA_xl)| 1.2B | 2048 | 24 | 1.70 
+
+
+# Usage 
+Instantiate a model like so:
+    from transformers import AutoModel, AutoModelForCausalLM
+    model = AutoModelForCausalLM.from_pretrained("Seledorn/RITA_s, trust_remote_code=True")
+    tokenizer = AutoTokenizer.from_pretrained("Seledorn/RITA_s")
+for generation use we support pipelines:
+   
+   
+    rita_gen = pipeline('text-generation', model=model, tokenizer = tokenizer)
+    sequences = rita_gen("MAB", max_length=20, do_sample=True, top_k=950, repetition_penalty=1.2, num_return_sequences=2, eos_token_id=2)
+    for seq in sequences:
+        print(f"seq: {seq['generated_text'].replace(' ', '')}")