Code Editor for app.py

import streamlit as st
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
from sqlalchemy import create_engine
from sqlalchemy.orm import Session
from sqlalchemy.orm import declarative_base

Base = declarative_base()

class StrategyData(Base):
    __tablename__ = 'strategy_data'
    id = Column(Integer, primary_key=True)
    stock = Column(String)
    date = Column(Date)
    price = Column(Float)
    signal = Column(Integer)

engine = create_engine('sqlite:///backtest.sqlite')
Base.metadata.create_all(bind=engine)

st.title('Finance Backtesting Engine')

# Upload stock data
uploaded_file = st.file_uploader('Upload your CSV file', type='csv')
if uploaded_file is not None:
    df = pd.read_csv(uploaded_file)
    st.write(df)

    # Simple moving average strategy
    df["SMA"] = df["Close"].rolling(window=20).mean()
    df["Signal"] = 0
    df["Signal"][20:] = np.where(df["Close"][20:] > df["SMA"][20:], 1, 0)
    df["Position"] = df["Signal"].diff()
    st.write(df)

    # Save strategy data to database
    with Session(engine) as session:
        for index, row in df.iterrows():
            strategy_data = StrategyData(stock=row['Stock'], date=row['Date'], price=row['Close'], signal=row['Signal'])
            session.add(strategy_data)
        session.commit()

    # Plot results
    plt.figure(figsize=(10,5))
    plt.plot(df['Close'], label='Close Price')
    plt.plot(df['SMA'], label='SMA 20', alpha=0.7)
    plt.plot(df[df['Position'] == 1].index, df['SMA'][df['Position'] == 1], '^', markersize=10, color='g', lw=0, label='Buy Signal')
    plt.plot(df[df['Position'] == -1].index, df['SMA'][df['Position'] == -1], 'v', markersize=10, color='r', lw=0, label='Sell Signal')
    plt.title('Strategy Backtesting')
    plt.legend() 
    st.pyplot()