Update app.py

app.py

@@ -8,7 +8,7 @@ tokenizer = AutoTokenizer.from_pretrained("OpenBuddy/openbuddy-openllama-3b-v10-
 model = AutoModelForCausalLM.from_pretrained("OpenBuddy/openbuddy-openllama-3b-v10-bf16")
 model.eval()
 
-with open('../system.prompt', 'r', encoding='utf-8') as f:
+with open('system.prompt', 'r', encoding='utf-8') as f:
     prompt = f.read()
 
 @app.post("/request")
@@ -16,7 +16,7 @@ async def echo():
     data = await request.get_json()
     if data.get("max_tokens") != None and data.get("max_tokens") > 500: data['max_tokens'] = 500
     userPrompt = prompt + "\n\nUser: " + data['request'] + "\nAssistant: "
-    input_ids = tokenizer.encode(prompt, return_tensors='pt')
+    input_ids = tokenizer.encode(userPrompt, return_tensors='pt')
     with torch.no_grad():
         output_ids = model.generate(
             input_ids=input_ids,