Spaces:
Runtime error
Runtime error
Upload 2 files
Browse files- app.py +78 -0
- requirements.txt +6 -0
app.py
ADDED
|
@@ -0,0 +1,78 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
from langchain_community.document_loaders import PyPDFLoader
|
| 2 |
+
from langchain_openai import OpenAIEmbeddings, ChatOpenAI
|
| 3 |
+
from langchain_community.embeddings import OllamaEmbeddings
|
| 4 |
+
from langchain_text_splitters import RecursiveCharacterTextSplitter
|
| 5 |
+
from langchain_community.vectorstores import Chroma
|
| 6 |
+
import os
|
| 7 |
+
from langchain.retrievers.multi_query import MultiQueryRetriever
|
| 8 |
+
from langchain_core.runnables import RunnablePassthrough
|
| 9 |
+
from langchain_core.output_parsers import StrOutputParser
|
| 10 |
+
from langchain.prompts import ChatPromptTemplate, PromptTemplate
|
| 11 |
+
import streamlit as st
|
| 12 |
+
|
| 13 |
+
os.environ["OPENAI_API_KEY"] =st.secrets["OPENAI_API_KEY"]
|
| 14 |
+
|
| 15 |
+
llm = ChatOpenAI(model='gpt-4o', temperature=0.2)
|
| 16 |
+
|
| 17 |
+
embeddings = OpenAIEmbeddings()
|
| 18 |
+
|
| 19 |
+
vector_store = Chroma(embedding_function=embeddings, persist_directory="mining-rag")
|
| 20 |
+
print("Vector store loaded successfully.")
|
| 21 |
+
|
| 22 |
+
|
| 23 |
+
question=st.text_input('whats your question?')
|
| 24 |
+
key=st.button('enter')
|
| 25 |
+
|
| 26 |
+
if key:
|
| 27 |
+
QUERY_PROMPT = PromptTemplate(
|
| 28 |
+
input_variables=["question"],
|
| 29 |
+
template="""You are an AI language model assistant. Your task is to generate three
|
| 30 |
+
different versions of the given user question to retrieve relevant documents from
|
| 31 |
+
a vector database. By generating multiple perspectives on the user question, your
|
| 32 |
+
goal is to help the user overcome some of the limitations of the distance-based
|
| 33 |
+
similarity search. Provide these alternative questions separated by newlines.
|
| 34 |
+
Original question: {question}""",
|
| 35 |
+
)
|
| 36 |
+
|
| 37 |
+
retriever = MultiQueryRetriever.from_llm(
|
| 38 |
+
vector_store.as_retriever(),
|
| 39 |
+
llm,
|
| 40 |
+
prompt=QUERY_PROMPT
|
| 41 |
+
)
|
| 42 |
+
|
| 43 |
+
WRITER_SYSTEM_PROMPT = "You are an AI critical thinker research assistant. Your sole purpose is to write well written, critically acclaimed, objective and structured reports on given text." # noqa: E501
|
| 44 |
+
# Report prompts from https://github.com/assafelovic/gpt-researcher/blob/master/gpt_researcher/master/prompts.py
|
| 45 |
+
RESEARCH_REPORT_TEMPLATE = """Information:
|
| 46 |
+
--------
|
| 47 |
+
{text}
|
| 48 |
+
--------
|
| 49 |
+
Using the above information, answer the following question or topic: "{question}" in a short manner-- \
|
| 50 |
+
The answer should focus on the answer to the question, should be well structured, informative, \
|
| 51 |
+
in depth, with facts and numbers if available and a minimum of 150 words and a maximum of 300 words.
|
| 52 |
+
You should strive to write the report using all relevant and necessary information provided.
|
| 53 |
+
You must write the report with markdown syntax.
|
| 54 |
+
You MUST determine your own concrete and valid opinion based on the given information. Do NOT deter to general and meaningless conclusions.
|
| 55 |
+
You must write the sources used in the context. if any article is used, mentioned in the end.
|
| 56 |
+
Please do your best, this is very important to my career.""" # noqa: E501
|
| 57 |
+
|
| 58 |
+
prompt = ChatPromptTemplate.from_messages(
|
| 59 |
+
[
|
| 60 |
+
("system", WRITER_SYSTEM_PROMPT),
|
| 61 |
+
("user", RESEARCH_REPORT_TEMPLATE),
|
| 62 |
+
]
|
| 63 |
+
)
|
| 64 |
+
|
| 65 |
+
chain = (
|
| 66 |
+
{"text": retriever, "question": RunnablePassthrough()}
|
| 67 |
+
| prompt
|
| 68 |
+
| llm
|
| 69 |
+
| StrOutputParser()
|
| 70 |
+
)
|
| 71 |
+
|
| 72 |
+
answer = chain.invoke(
|
| 73 |
+
{
|
| 74 |
+
"question": question
|
| 75 |
+
}
|
| 76 |
+
)
|
| 77 |
+
st.write(answer)
|
| 78 |
+
|
requirements.txt
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
langchain==0.2.0
|
| 2 |
+
langchain_community==0.2.0
|
| 3 |
+
langchain_core==0.2.0
|
| 4 |
+
langchain_openai==0.1.7
|
| 5 |
+
langchain_text_splitters==0.2.0
|
| 6 |
+
streamlit==1.33.0
|