app.py

import pandas as pd

splits = {'train': 'data/train-00000-of-00001.parquet', 'test': 'data/test-00000-of-00001.parquet'} 
df = pd.read_parquet("hf://datasets/pierre-pessarossi/climate-question-answers/" + splits["train"]) 
df = df.sample(500) 
 
context_data = []
for i in range(len(df)):
    context = ""
    for j in range(2):
        column_name = df.columns[j]
        value = df.iloc[i][j]
        # Handle NoneType values
        if value is None:
            value = ""
        context += f"{column_name}: {value} "
    context_data.append(context)


import os

# Get the secret key from the environment
groq_key = os.environ.get('groq_api_keys')

## LLM used for RAG
from langchain_groq import ChatGroq

llm = ChatGroq(model="llama-3.1-70b-versatile",api_key=groq_key)

## Embedding model!
from langchain_huggingface import HuggingFaceEmbeddings
embed_model = HuggingFaceEmbeddings(model_name="mixedbread-ai/mxbai-embed-large-v1")

# create vector store!
from langchain_chroma import Chroma

vectorstore = Chroma(
    collection_name="climate_education",
    embedding_function=embed_model,
    persist_directory="./",
)

# add data to vector nstore
vectorstore.add_texts(context_data)

retriever = vectorstore.as_retriever()

from langchain_core.prompts import PromptTemplate

template = ("""You are a simplified climate education expert.
    Use the provided context to answer the question.
    If you don't know the answer, say so. Explain your answer in detail.
    Do not discuss the context in your response; just provide the answer directly.
    Context: {context}
    Question: {question}
    Answer:""")

rag_prompt = PromptTemplate.from_template(template)

from langchain_core.output_parsers import StrOutputParser
from langchain_core.runnables import RunnablePassthrough

rag_chain = (
    {"context": retriever, "question": RunnablePassthrough()}
    | rag_prompt
    | llm
    | StrOutputParser()
)

import gradio as gr

def rag_memory_stream(message, history):
    partial_text = ""
    for new_text in rag_chain.stream(message):
        partial_text += new_text
        yield partial_text

examples = ['What is climate change', 'What is carbon emission', 'What are the indicators of extreme climates', 'What happens during the completion stage of an oil well?']


title = "Climate Education Buddy :)"
description = "Real-time AI App with Groq API and LangChain to Provide Simplified Climate-related Answers..."
# demo = gr.Interface(
#     title=title,
#     fn=rag_memory_stream,
#     #inputs="text",
#     inputs=gr.Textbox(
#         label="Enter Your Question",
#         placeholder="E.g., What is gas emission?"),
#     #outputs="text",
#     outputs=gr.Textbox(
#         label="AI Response",
#         placeholder="E.g., Gas emissions refer to the release of gases into the atmosphere that contribute to the greenhouse effect and global warming"),
#     examples=examples,
#     theme="citrus",
#     allow_flagging="never", )

demo = gr.ChatInterface(fn=rag_memory_stream,
                        type="messages",
                        title=title,
                        description=description,
                        fill_height=True,
                        examples=examples,
                        theme="glass",
                        )
if __name__ == "__main__":
    demo.launch()