import streamlit as st import pandas as pd import numpy as np import io import matplotlib.pyplot as plt from sklearn.preprocessing import LabelEncoder import seaborn as sns import base64 import json from langchain.docstore.document import Document from langchain.vectorstores import Chroma from langchain.embeddings import HuggingFaceEmbeddings from langchain.llms import HuggingFaceHub from langchain.chains import RetrievalQA from Information import show_general_data_statistics, describe_data, info_data from Preprocessing1 import preview_data, data_cleaning, modify_column_names from Preprocessing2 import handle_categorical_values, missing_values, handle_duplicates, handle_outliers from RAG import create_doucment, ask_me, load_models_embedding, load_models_llm, create_database from langchain.vectorstores import FAISS # Helper Functions def create_documents(df): """Converts a DataFrame into a list of Document objects.""" documents = [ Document( metadata={"id": str(i)}, page_content=json.dumps(row.to_dict()) ) for i, row in df.iterrows() ] return documents def load_embedding_model(): """Loads the embedding model for vectorization.""" return HuggingFaceEmbeddings(model_name="sentence-transformers/all-MiniLM-L6-v2") def load_llm(api_key): """Loads the LLM model for answering queries.""" return HuggingFaceHub( repo_id="mistralai/Mixtral-8x7B-Instruct-v0.1", huggingfacehub_api_token=api_key, model_kwargs={"temperature": 0.5, "max_length": 100} ) def ask_question(question, retriever, llm): """Uses a QA chain to retrieve and answer a question.""" qa_chain = RetrievalQA.from_chain_type( retriever=retriever, chain_type="stuff", llm=llm, return_source_documents=False ) response = qa_chain.invoke({"query": question}) return response["result"] # Streamlit App def upload_data(): st.title("Upload Dataset") file = st.file_uploader("Upload your dataset", type=["csv", "xlsx"]) if file: try: if file.name.endswith(".csv"): data = pd.read_csv(file) elif file.name.endswith(".xlsx"): data = pd.read_excel(file) st.session_state["data"] = data st.success("Dataset uploaded successfully!") except Exception as e: st.error(f"Error loading file: {e}") def preview_data(): if "data" in st.session_state: st.title("Preview Dataset") st.dataframe(st.session_state["data"]) else: st.warning("Please upload a dataset first.") api="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21" def rag_chatbot(): st.title("RAG Chatbot") # Check if data is uploaded if "data" in st.session_state and isinstance(st.session_state["data"], pd.DataFrame): df = st.session_state["data"] # Convert data to documents st.write("Processing the dataset...") documents = create_documents(df) # Load models st.write("Loading models...") embedding_model = load_embedding_model() llm_model = load_llm(api_key=api[:-2]) # Create retriever using Chroma # FAISS.from_documents(documents, embedding) retriever = FAISS.from_documents(documents, embedding=embedding_model).as_retriever() # Ask a question question = st.text_input("Ask a question about your dataset:") if question: response = ask_question(question, retriever, llm_model) st.write(f"Answer: {response}") else: st.warning("Please upload a dataset to proceed.") def main(): st.sidebar.title("Navigation") options = st.sidebar.radio( "Go to", ["Upload", "Preview", "RAG Chatbot"], key="navigation_key" ) if options == "Upload": upload_data() elif options == "Preview": preview_data() elif options == "RAG Chatbot": rag_chatbot() if __name__ == "__main__": main() # def upload_data(): # st.title("Upload Dataset") # file = st.file_uploader("Upload your dataset", type=[ # "csv", "xlsx"], key="file_uploader_1") # if file: # try: # if file.name.endswith(".csv"): # data = pd.read_csv(file) # elif file.name.endswith(".xlsx"): # data = pd.read_excel(file) # st.session_state["data"] = data # st.success("Dataset uploaded successfully!") # except Exception as e: # st.error(f"Error loading file: {e}") # return file # def download_data(): # """Downloads the DataFrame as a CSV file.""" # if "data" in st.session_state and not st.session_state["data"].empty: # csv = st.session_state["data"].to_csv(index=False).encode('utf-8') # st.download_button( # label="Download Cleaned Dataset", # data=csv, # file_name="cleaned_data.csv", # mime="text/csv" # ) # else: # st.warning( # "No data available to download. Please modify or upload a dataset first.") # def rag_chatbot(): # st.title("RAG Chatbot") # # Check if data is uploaded # if "data" in st.session_state and isinstance(st.session_state["data"], pd.DataFrame): # df = st.session_state["data"] # # Convert data to documents # st.write("Processing the dataset...") # documents = create_doucment(df) # st.write(f"Created {len(documents)} documents.") # # Load models # st.write("Loading models...") # embedding = load_models_embedding() # llm = load_models_llm() # # Create retriever # retriever = create_database(embedding, documents).as_retriever() # # Ask a question # question = st.text_input("Ask a question about your dataset:") # if question: # response = ask_me(question, retriever, llm) # st.write(f"Answer: {response}") # else: # st.warning("Please upload a dataset to proceed.") # def main(): # st.sidebar.title("Navigation") # options = st.sidebar.radio( # "Go to", # [ # "Upload", # "Preview", # "Data Cleaning", # "Modify Column Names", # "General Data Statistics", # "Describe", # "Info", # "Handle Categorical", # "Missing Values", # "Handle Duplicates", # "Handle Outliers", # "Download", # "RAG Chatbot" # ], # key="unique_navigation_key", # ) # if options == "Upload": # upload_data() # elif options == "Preview": # preview_data() # elif options == "Data Cleaning": # data_cleaning() # elif options == "Modify Column Names": # modify_column_names() # elif options == "General Data Statistics": # show_general_data_statistics() # elif options == "Describe": # describe_data() # elif options == "Info": # info_data() # elif options == "Handle Categorical": # handle_categorical_values() # elif options == "Missing Values": # missing_values() # elif options == "Handle Duplicates": # handle_duplicates() # elif options == "Handle Outliers": # handle_outliers() # elif options == "Download": # download_data() # elif options == "RAG Chatbot": # rag_chatbot() # else: # st.warning("Please upload a dataset first.") # if __name__ == "__main__": # main()