adding HF pipeline for llm model

app.py

@@ -35,8 +35,8 @@ def clear_chat():
     return []
 
 def create_prompt():
-    prompt_template = """Asnwer the questions regarding the conversation in the Audio . 
-    Use the following context to answer. 
+    prompt_template = """You are a chatbot that answers questions regarding the conversation in given context . 
+    Use the following context to answer in sentences and points. 
     If you don't know the answer, just say I don't know. 
 
     {context}
@@ -78,14 +78,7 @@ def audio_processor(wav_file,API_key,wav_model='small',llm='HuggingFace',tempera
     global qa
 
     if llm == 'HuggingFace':
-        chat = llm_ops.get_hugging_face_model(
-                            model_id="VMware/open-llama-7B-open-instruct",
-                            API_key=API_key,
-                            temperature=temperature,
-                            max_tokens=max_tokens
-                            )
-    else:
-        chat = llm_ops.get_openai_chat_model(API_key=API_key)
+        chat = llm_ops.get_llama_model()
     
     chain_type_kwargs = {"prompt": create_prompt()}
     qa = RetrievalQA.from_chain_type(llm=chat,
@@ -133,7 +126,7 @@ css="""
 
 title = """
 <div style="text-align: center;max-width: 2048px;">
-    <h1>Q&A using LLAMA on Audio files</h1>
+    <h1>Q&A with LLAMA on Audio files</h1>
     <p style="text-align: center;">Upload a Audio file/link and  query  LLAMA-chatbot.
     <i> Tools uses State of the Art Models from  HuggingFace/OpenAI so, make sure to add your key.</i>
     </p>

llm_ops.py

@@ -1,4 +1,8 @@
 import os 
+import torch 
+from langchain import HuggingFacePipeline
+from transformers import AutoTokenizer
+import transformers
 
 def get_openai_chat_model(API_key):
     try:
@@ -9,13 +13,20 @@ def get_openai_chat_model(API_key):
     llm = OpenAI()
     return llm
 
-def get_hugging_face_model(model_id,API_key,temperature=0.1,max_tokens=4096):
-    try:
-        from langchain import HuggingFaceHub
-    except ImportError as err:
-        raise "{}, unable to load openAI. Please install openai and add OPENAIAPI_KEY"
-    chat_llm = HuggingFaceHub(huggingfacehub_api_token=API_key,
-                                      repo_id=model_id,
-                                      model_kwargs={"temperature": temperature, "max_new_tokens": max_tokens})
-    return chat_llm
 
+def get_llama_model():
+    model = "meta-llama/Llama-2-7b-chat-hf"
+
+    tokenizer = AutoTokenizer.from_pretrained(model)
+
+    pipeline = transformers.pipeline("text-generation", 
+        model=model,
+        tokenizer=tokenizer,
+        torch_dtype=torch.bfloat16,
+        trust_remote_code=True,
+        device_map="auto",
+        max_length=1000,
+        eos_token_id=tokenizer.eos_token_id
+        )
+    llm = HuggingFacePipeline(pipeline = pipeline, model_kwargs = {'temperature':0})
+    return llm