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=1024)
+pipe_flan = pipeline("text2text-generation", model="google/flan-t5-small", max_new_tokens=500)
 
 app = FastAPI()
 
@@ -32,7 +32,10 @@ def llama():
     # print(tresponse)
     input = "Generate 5 jokes"
     output = pipe_flan(input)
-    response_message = {"message": output[0]["generated_text"]}
+    text = ""
+    for o in output:
+        text = text + ' ' + o["generated_text"]
+    response_message = {"message": text}
     json_response = jsonable_encoder(response_message)
     return JSONResponse(content=json_response)