app.py

import streamlit as st
import pandas as pd
import json
from langchain import OpenAI
from langchain_experimental.agents.agent_toolkits.pandas.base import create_pandas_dataframe_agent
import pandas as pd


API_KEY = st.secrets["API_KEY"]


page_element="""
<style>
[data-testid="stAppViewContainer"]{
background-image: url("https://www.surveylegend.com/wordpress/wp-content/themes/sage/resources/images/img/modal/upgrade-file-upload.png");
background-size: cover;
}
[data-testid="stHeader"]{
background-color: rgba(0,0,0,0);
}
[data-testid="stToolbar"]{
right: 2rem;
background-image: url("");
background-size: cover;
}
[data-testid="stSidebar"]> div:first-child{
background-image: url("https://img.freepik.com/premium-vector/skyblue-gradient-background-advertisers-gradient-hq-wallpaper_189959-513.jpg");
background-size: cover;
}
</style>

"""
st.markdown(page_element, unsafe_allow_html=True)
st.markdown("<h1 style='text-align: center; color: white';>Dynamic Analytics Crime Modelling 🚔👮</h1>", unsafe_allow_html=True)
st.markdown("---")


# st.write("Please upload your Chicago Crime CSV file")

# data = st.file_uploader("Upload a CSV")

file_path = "Crimes-2023.csv"  
df = pd.read_csv(file_path)
df=df.dropna()
with st.sidebar:
    st.title("Dataset Description")
    st.write("This dataset contains information about crimes in the year 2023.")
    st.write("It includes various attributes such as ID, Case Number, Date, Block, Primary Type, Location Description, Arrest, and more.")

    st.subheader("Dataset Features")
    crime_counts = df.columns
    result = ", ".join(crime_counts)
    st.write(result)


    st.write(f"The dataset consists of {len(df)} records.")

    st.header("Exploring Crime Dataset")
    st.subheader("Dataset Overview")
    st.write("Here's an overview of the dataset:")
    st.dataframe(df.head())

    st.subheader("Dataset Statistics")
    st.write("Here are some statistics about the dataset:")
    st.write(df.describe())

    st.subheader("Crime Types")
    crime_counts = df["Primary Type"].value_counts()
    st.bar_chart(crime_counts)


query = st.text_input("Please do ask here to get insights from the CSV - ")

def agent(df):
    llm = OpenAI(openai_api_key=API_KEY)

    return create_pandas_dataframe_agent(llm, df, verbose=False)


def query_llm(agent, query):

    prompt = (
        """
            For the following query, if it requires drawing a table, reply as follows:
            {"table": {"columns": ["column1", "column2", ...], "data": [[value1, value2, ...], [value1, value2, ...], ...]}}

            If the query requires creating a bar chart, reply as follows:
            {"bar": {"columns": ["A", "B", "C", ...], "data": [25, 24, 10, ...]}}

            If the query requires creating a line chart, reply as follows:
            {"line": {"columns": ["A", "B", "C", ...], "data": [25, 24, 10, ...]}}

            The various types of chart, "bar", "line" and "pie" plot

            If it is just asking a question that requires neither, reply as follows:
            {"answer": "answer"}
            Example:
            {"answer": "The title with the highest rating is 'Gilead'"}

            If you do not know the answer, reply as follows:
            {"answer": "I do not know."}

            Return all output as a string.

            All strings in "columns" list and data list, should be in double quotes,

            For example: {"columns": ["title", "ratings_count"], "data": [["Gilead", 361], ["Spider's Web", 5164]]}

            Lets think step by step.

            Below is the query.
            Query: 
            """
        + query
    )

    response = agent.run(prompt)
    return response.__str__()



def query_analysis(agent):

    prompt = (
        """

            If you do not know the answer, reply as follows:
            {"answer": "I do not know."}

            Return all output as a string.

            All strings in "columns" list and data list, should be in double quotes,

            For example: {"columns": ["title", "ratings_count"], "data": [["Gilead", 361], ["Spider's Web", 5164]]}

            Lets think step by step.

            Below is the query.
            Query:Prepare the analytical questions by looking the complex realtion, trend and behavior of the various columns in the csv file

            The Response should be multiple questions all in one string
            Response: 
            """
    )

    response = str(agent.run(prompt))
    return response.__str__()




def decode_response(response: str) -> dict:

    return json.loads(response)


def write_response(response_dict: dict):

    if "answer" in response_dict:
        st.write(response_dict["answer"])

    if "bar" in response_dict:
        data = response_dict["bar"]
        df = pd.DataFrame(data)
        df.set_index("columns", inplace=True)
        st.bar_chart(df)

    if "line" in response_dict:
        data = response_dict["line"]
        df = pd.DataFrame(data)
        df.set_index("columns", inplace=True)
        st.line_chart(df)

    if "table" in response_dict:
        data = response_dict["table"]
        df = pd.DataFrame(data["data"], columns=data["columns"])
        st.table(df)

col_1, col_2= st.columns(2)

with col_1:
    # if st.button("Get Question", type="primary"):
    #             agent = agent(df)
    #             response = query_analysis(agent=agent)
    #             decoded_response = decode_response(response)
    #             write_response(decoded_response)
    pass

with col_2:
    if st.button("Submit Query", type="primary"):

        agent = agent(df)
        response = query_llm(agent=agent, query=query)
        decoded_response = decode_response(response)
        write_response(decoded_response)