Spaces:
Running
Running
File size: 2,640 Bytes
fe5256f dbc8314 fe5256f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 |
from langchain.callbacks.manager import CallbackManager
from langchain.callbacks.streaming_stdout import StreamingStdOutCallbackHandler
from langchain import PromptTemplate
from langchain_community.llms import LlamaCpp
#from langchain.chains import RetrievalQA
#from langchain_community.embeddings import SentenceTransformerEmbeddings
from langchain_core.prompts import ChatPromptTemplate
from langchain.callbacks.base import BaseCallbackHandler
#from langchain.schema import HumanMessage
import os
import json,streamlit as st
from pathlib import Path
class StreamHandler(BaseCallbackHandler):
def __init__(self, container, initial_text=""):
self.container = container
self.text=initial_text
def on_llm_new_token(self, token: str, **kwargs) -> None:
# "/" is a marker to show difference
# you don't need it
#self.text+=token+"/"
self.text+=token
self.container.markdown(self.text)
st.title("Prompt Engineer")
# Main chat form
with st.form("chat_form"):
query = st.text_input("Enter the topic you want to generate prompt for?: ")
#LLM_Summary = st.checkbox('Summarize results with LLM')
submit_button = st.form_submit_button("Send")
template = """
<s>[INST] <<SYS>>
Act as a patent advisor by providing subject matter expertise on any topic. Provide detailed and elaborate answers
<</SYS>>
{text} [/INST]
"""
response=""
prompt = PromptTemplate(
input_variables=["text"],
template=template,
)
text = "Help me create a good prompt for the following: Information that is needed to file a US patent application for " + query
#print(prompt.format(text=query))
# Callbacks support token-wise streaming
callback_manager = CallbackManager([StreamingStdOutCallbackHandler()])
#model_path = "C:\Rajesh\AI-ML-Training\LLM\llama-2-7b.Q4_K_M.gguf"\
model_path = "zephyr-7b-beta.Q4_K_M.gguf"
chat_box=st.empty()
stream_handler = StreamHandler(chat_box)
llm = LlamaCpp(
model_path=model_path,
temperature=0.8,
max_tokens=500,
top_p=1,
#streaming=True,
#callback_manager=callback_manager,
callback_manager = [stream_handler],
verbose=True, # Verbose is required to pass to the callback manager
)
if submit_button:
#st.write("Fetching results..\n")
output = llm.invoke(prompt.format(text=text))
#response = response+output
#st.write(response)
#response = output([HumanMessage(content=query)])
#llm_response = output.content
#st.markdown(output)
|