flan

question_paper.py

@@ -4,7 +4,7 @@ from fastapi.middleware.cors import CORSMiddleware
 from fastapi.encoders import jsonable_encoder
 from fastapi.responses import JSONResponse
 
-pipe_flan = pipeline("text2text-generation", model="google/flan-t5-small", max_new_tokens=500)
+pipe_flan = pipeline("text2text-generation", model="google/flan-t5-small", max_new_tokens=100)
 
 app = FastAPI()
 
@@ -22,7 +22,7 @@ app.add_middleware(
 )
 
 @app.get("/")
-def llama():
+def llama(input):
     # prompt = [{'role': 'user', 'content': ""+input}]
     # inputs = TOKENIZER.apply_chat_template( prompt, add_generation_prompt=True,     return_tensors='pt' )
 
@@ -30,7 +30,6 @@ def llama():
 
     # tresponse = TOKENIZER.decode(tokens[0], skip_special_tokens=False)
     # print(tresponse)
-    input = "Generate 5 jokes"
     output = pipe_flan(input)
     text = ""
     for o in output: