app.py

from datasets import load_dataset
dataset = load_dataset("Namitg02/Test")
print(dataset)

from langchain.docstore.document import Document as LangchainDocument

#RAW_KNOWLEDGE_BASE = [LangchainDocument(page_content=["dataset"])]

from langchain.text_splitter import RecursiveCharacterTextSplitter
splitter = RecursiveCharacterTextSplitter(chunk_size=100, chunk_overlap=15,separators=["\n\n", "\n", " ", ""])
#docs = splitter.split_documents(RAW_KNOWLEDGE_BASE)
docs = splitter.create_documents(str(dataset))


from langchain_community.embeddings import HuggingFaceEmbeddings
embedding_model = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
# embeddings = embedding_model.encode(docs)


from langchain_community.vectorstores import Chroma
persist_directory = 'docs/chroma/'

vectordb = Chroma.from_documents(
    documents=docs,
    embedding=embedding_model,
    persist_directory=persist_directory
)

#docs_ss = vectordb.similarity_search(question,k=3)

# Create placeholders for the login form widgets using st.empty()
#user_input_placeholder = st.empty()
#pass_input_placeholder = st.empty()

#from langchain_community.output_parsers.rail_parser import GuardrailsOutputParser
#from langchain.prompts import PromptTemplate

#template = """Use the following pieces of context to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer. Use three sentences maximum. Keep the answer as concise as possible. Always say "thanks for asking!" at the end of the answer. 
#{You are a helpful dietician}
#Question: {question}
#Helpful Answer:"""

#QA_CHAIN_PROMPT = PromptTemplate.from_template(template)

#from langchain.chains import ConversationalRetrievalChain
#from langchain.memory import ConversationBufferMemory
#memory = ConversationBufferMemory(
#    memory_key="chat_history",
#    return_messages=True
#)

question = "How can I reverse Diabetes?"
#print("template")

retriever = vectordb.as_retriever(
    search_type="similarity", search_kwargs={"k": 2}
)


from transformers import pipeline
from langchain_community.llms.huggingface_pipeline import HuggingFacePipeline
from langchain_core.messages import SystemMessage
from langchain_core.prompts import HumanMessagePromptTemplate
from langchain_core.prompts import ChatPromptTemplate
print("check1")

qa_chat_prompt = ChatPromptTemplate.from_messages(
[
        SystemMessage(
            content=(
                "You are a Diabetes eductaor that provide advice to patients."
            )
        ),
        HumanMessagePromptTemplate.from_template("{context}"),
    ]
)

llm_model = "deepset/roberta-base-squad2"
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained(llm_model)
#question = "How can I reverse diabetes?"

print("check2")
pipe = pipeline(model = llm_model, tokenizer = tokenizer, task = "question-answering", temperature=0.2)
#"question-answering", "conversational"

print("check3")
chain = pipe(question = "How can I reverse diabetes?",context = "You should reply like a doctor") 
#(question = question, context = context)
print("check3A")
import gradio as gr
#ragdemo = gr.load("models/HuggingFaceH4/zephyr-7b-beta")
ragdemo = gr.Interface.from_pipeline(chain)
print(chain)
print("check4")
ragdemo.launch()
print("check5")