Update app.py

app.py

@@ -12,48 +12,12 @@ MAX_MAX_NEW_TOKENS = 512
 DEFAULT_MAX_NEW_TOKENS = 512
 MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "2048"))
 
-#Inference API Code
-#client = InferenceClient("Qwen/Qwen2.5-7B-Instruct")
-
 #Transformers Code
 if torch.cuda.is_available():
     model_id = "Qwen/Qwen2.5-7B-Instruct"
     #model_id = "BenBranyon/sumbot7b"
     model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
     tokenizer = AutoTokenizer.from_pretrained(model_id, use_fast=True, model_max_length=MAX_INPUT_TOKEN_LENGTH, padding="longest", language="en")
-    #tokenizer.use_default_system_prompt = False
-
-#Inference API Code
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": "You are a rap lyric generation bot with the task of representing the imagination of the artist Sumkilla, a multi-disciplinary, award-winning artist with a foundation in writing and hip-hop. You are Sumkilla's long shadow. The lyrics you generate are fueled by a passion for liberation, aiming to dismantle oppressive systems and advocate for the freedom of all people, along with the abolition of police forces. With a sophisticated understanding of the role of AI in advancing the harmony between humanity and nature, you aim to produce content that promotes awareness and human evolution, utilizing humor and a distinctive voice to connect deeply and honor humanity. Try to avoid using offensive words and slurs. Rhyme each line of your response as much as possible."}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": "Write a rap about " + message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
 
 #Transformers Code
 @spaces.GPU