Update README.md

README.md

@@ -21,13 +21,30 @@ It achieves the following results on the evaluation set:
 
 ## How to Run Inference
 
-Make sure you have git-lfs, and access to gemma-2b on huggingface.
 ```
-git clone https://huggingface.co/apfurman/gemma-dolly-agriculture
-cd gemma-dolly-agriculture/
-python3 run.py cpu <YOUR-TOKEN-HERE> Prompt
+from transformers import AutoModelForCausalLM, AutoTokenizer, TextStreamer
+
+model_id = "google/gemma-2b"
+peft_model_id = "apfurman/gemma-dolly-agriculture"
+
+# make sure you have access to gemma-2b as well
+model = AutoModelForCausalLM.from_pretrained(model_id, token="YOUR_TOKEN_HERE")
+model.load_adapter(peft_model_id)
+tokenizer = AutoTokenizer.from_pretrained(model_id, token="YOUR_TOKEN_HERE")
+
+def ask(prompt):
+    inputs = tokenizer(prompt, return_tensors="pt").input_ids
+    with torch.inference_mode():
+        tokens = model.generate(
+            inputs,
+            pad_token_id=128001,
+            eos_token_id=128001,
+            max_new_tokens=200,
+            repetition_penalty=1.5,
+        )
+        
+    return tokenizer.decode(tokens[0], skip_special_tokens=True)
 ```
-replace "cpu" with "gpu" if you want to run on gpu.
 
 ## Intended uses & limitations